{"updated":"2026-09-14T02:46:23.470Z","boards":[{"name":"SWE-Bench Pro","slug":"swe-bench-pro","metric":"resolve rate","entries":[{"model":"anthropic/claude-fable-5","lab":"anthropic","score":80.3,"best_input_per_m":3,"best_output_per_m":18.5,"points_per_dollar":721.99},{"model":"anthropic/claude-opus-5","lab":"anthropic","score":79.2,"best_input_per_m":5,"best_output_per_m":25,"points_per_dollar":499.84},{"model":"sakana/fugu-ultra","lab":"sakana","score":73.7,"best_input_per_m":5,"best_output_per_m":30,"points_per_dollar":406.73},{"model":"anthropic/claude-opus-4-8","lab":"anthropic","score":69.2,"best_input_per_m":0.425,"best_output_per_m":2.125,"points_per_dollar":4196.48},{"model":"alibaba/qwen3.8-max-preview","lab":"alibaba","score":67.7,"best_input_per_m":2,"best_output_per_m":6,"points_per_dollar":1578.09},{"model":"xai/grok-4.5","lab":"xai","score":64.7,"best_input_per_m":2,"best_output_per_m":6,"points_per_dollar":1493.54},{"model":"openai/gpt-5.6-sol","lab":"openai","score":64.6,"best_input_per_m":2,"best_output_per_m":10,"points_per_dollar":1019.25},{"model":"openai/gpt-5.6-terra","lab":"openai","score":63.4,"best_input_per_m":1.5,"best_output_per_m":2,"points_per_dollar":3274.79},{"model":"anthropic/claude-sonnet-5","lab":"anthropic","score":63.2,"best_input_per_m":1.44,"best_output_per_m":7.2,"points_per_dollar":1131.16},{"model":"openai/gpt-5.6-luna","lab":"openai","score":62.7,"best_input_per_m":0.06,"best_output_per_m":0.37,"points_per_dollar":23412.99},{"model":"zhipuai/glm-5.2","lab":"zhipuai","score":62.1,"best_input_per_m":0.3,"best_output_per_m":1.05,"points_per_dollar":8518.52},{"model":"alibaba/qwen3.8-27b","lab":"alibaba","score":61.7,"best_input_per_m":0.1,"best_output_per_m":0.4,"points_per_dollar":23513.72},{"model":"meta/muse-spark-1.1","lab":"meta","score":61.5,"best_input_per_m":1.25,"best_output_per_m":4.25,"points_per_dollar":2101.85},{"model":"alibaba/qwen3.7-max","lab":"alibaba","score":60.6,"best_input_per_m":0.825,"best_output_per_m":2.4755,"points_per_dollar":3326.29},{"model":"meituan/longcat-2.0","lab":"meituan","score":59.5,"best_input_per_m":0.3,"best_output_per_m":1.2,"points_per_dollar":7757.09},{"model":"openai/gpt-5.4","lab":"openai","score":59.1,"best_input_per_m":0.75,"best_output_per_m":6,"points_per_dollar":1704.15},{"model":"minimax/MiniMax-M3","lab":"minimax","score":59,"best_input_per_m":0.225,"best_output_per_m":0.9,"points_per_dollar":9683.24},{"model":"sakana/fugu","lab":"sakana","score":59,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null},{"model":"google/gemini-3.6-flash","lab":"google","score":58.7,"best_input_per_m":0.375,"best_output_per_m":1.875,"points_per_dollar":5299.03},{"model":"xiaomi/mimo-v2.5-pro","lab":"xiaomi","score":57.2,"best_input_per_m":0.4,"best_output_per_m":0.8,"points_per_dollar":9247.88},{"model":"stepfun/step-3.7-flash","lab":"stepfun","score":56.3,"best_input_per_m":0.185,"best_output_per_m":1.11,"points_per_dollar":8207.24},{"model":"minimax/MiniMax-M2.7","lab":"minimax","score":56.2,"best_input_per_m":0.08,"best_output_per_m":0.32,"points_per_dollar":25841.46},{"model":"google/gemini-3.5-flash","lab":"google","score":55.1,"best_input_per_m":0.1857,"best_output_per_m":1.1142,"points_per_dollar":6774.32},{"model":"openai/gpt-5.4-mini","lab":"openai","score":54.4,"best_input_per_m":0.375,"best_output_per_m":4,"points_per_dollar":2435.09},{"model":"google/gemini-3.5-flash-lite","lab":"google","score":54.2,"best_input_per_m":0.15,"best_output_per_m":1.25,"points_per_dollar":7752.82},{"model":"openai/gpt-5.4-nano","lab":"openai","score":52.4,"best_input_per_m":0.18,"best_output_per_m":1.1,"points_per_dollar":7911.58},{"model":"anthropic/claude-opus-4-6","lab":"anthropic","score":51.9,"best_input_per_m":4,"best_output_per_m":20,"points_per_dollar":777.41},{"model":"meta/muse-glimmer-30b","lab":"meta","score":51.2,"best_input_per_m":0.2,"best_output_per_m":0.8,"points_per_dollar":9309.09},{"model":"microsoft/mai-code-1-flash","lab":"microsoft","score":51.2,"best_input_per_m":0.75,"best_output_per_m":4.5,"points_per_dollar":1883.74},{"model":"poolside/laguna-xs-2.1","lab":"poolside","score":47.6,"best_input_per_m":0.06,"best_output_per_m":0.12,"points_per_dollar":40476.19},{"model":"google/gemini-3.1-pro-preview","lab":"google","score":46.1,"best_input_per_m":1,"best_output_per_m":6,"points_per_dollar":1325.57},{"model":"anthropic/claude-opus-4-5-20251101","lab":"anthropic","score":45.89,"best_input_per_m":4.5,"best_output_per_m":22.5,"points_per_dollar":262.83},{"model":"anthropic/claude-opus-4-7","lab":"anthropic","score":44.9,"best_input_per_m":4,"best_output_per_m":20,"points_per_dollar":672.56},{"model":"anthropic/claude-sonnet-4-5","lab":"anthropic","score":43.6,"best_input_per_m":0.43,"best_output_per_m":2.14,"points_per_dollar":2621.14},{"model":"google/gemini-3-pro-preview","lab":"google","score":43.3,"best_input_per_m":0.57,"best_output_per_m":3.43,"points_per_dollar":1730.89},{"model":"openai/gpt-5","lab":"openai","score":41.78,"best_input_per_m":0.2,"best_output_per_m":1.2,"points_per_dollar":5693.65},{"model":"openai/gpt-5.2-codex","lab":"openai","score":41.04,"best_input_per_m":0.14,"best_output_per_m":1.14,"points_per_dollar":5377.36},{"model":"cohere/north-mini-code-1-0","lab":"cohere","score":40.2,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null},{"model":"anthropic/claude-haiku-4-5","lab":"anthropic","score":39.45,"best_input_per_m":0.14,"best_output_per_m":0.71,"points_per_dollar":7196.28},{"model":"alibaba/qwen3-coder-480b-a35b-instruct","lab":"alibaba","score":38.7,"best_input_per_m":0.2,"best_output_per_m":0.8,"points_per_dollar":5724.85},{"model":"minimax/MiniMax-M2.1","lab":"minimax","score":36.81,"best_input_per_m":0.27,"best_output_per_m":1.1,"points_per_dollar":3989.81},{"model":"google/gemini-3-flash-preview","lab":"google","score":34.63,"best_input_per_m":0.07,"best_output_per_m":0.43,"points_per_dollar":11113.61},{"model":"openai/gpt-5.5","lab":"openai","score":30.9,"best_input_per_m":0.1875,"best_output_per_m":1.125,"points_per_dollar":3762.56},{"model":"openai/gpt-5.2","lab":"openai","score":29.94,"best_input_per_m":0.25,"best_output_per_m":2,"points_per_dollar":2226.02},{"model":"moonshotai/kimi-k2.6","lab":"moonshotai","score":27.3,"best_input_per_m":0.275,"best_output_per_m":1.1,"points_per_dollar":2937.06},{"model":"alibaba/qwen3-235b-a22b","lab":"alibaba","score":21.41,"best_input_per_m":0.3,"best_output_per_m":0.5,"points_per_dollar":3979.55},{"model":"zhipuai/glm-5.1","lab":"zhipuai","score":19.8,"best_input_per_m":0.45,"best_output_per_m":2.15,"points_per_dollar":1518.17},{"model":"deepseek/deepseek-v4-pro","lab":"deepseek","score":18,"best_input_per_m":0.35,"best_output_per_m":0.8,"points_per_dollar":3043.01},{"model":"anthropic/claude-sonnet-4-6","lab":"anthropic","score":14.9,"best_input_per_m":0.9,"best_output_per_m":5.55,"points_per_dollar":446.56},{"model":"zhipuai/glm-4.6","lab":"zhipuai","score":9.67,"best_input_per_m":0.286,"best_output_per_m":1.142,"points_per_dollar":1001.37},{"model":"meta/llama-4-maverick-17b-instruct","lab":"meta","score":5.24,"best_input_per_m":0.14,"best_output_per_m":0.59,"points_per_dollar":1073.33}]},{"name":"SWE-Bench Verified","slug":"swe-bench-verified","metric":"resolved","entries":[{"model":"anthropic/claude-opus-5","lab":"anthropic","score":96,"best_input_per_m":5,"best_output_per_m":25,"points_per_dollar":605.87},{"model":"anthropic/claude-fable-5","lab":"anthropic","score":95,"best_input_per_m":3,"best_output_per_m":18.5,"points_per_dollar":854.16},{"model":"anthropic/claude-opus-4-8","lab":"anthropic","score":88.6,"best_input_per_m":0.425,"best_output_per_m":2.125,"points_per_dollar":5372.95},{"model":"anthropic/claude-sonnet-5","lab":"anthropic","score":85.2,"best_input_per_m":1.44,"best_output_per_m":7.2,"points_per_dollar":1524.91},{"model":"deepseek/deepseek-v4-pro","lab":"deepseek","score":80.6,"best_input_per_m":0.35,"best_output_per_m":0.8,"points_per_dollar":13625.91},{"model":"minimax/MiniMax-M3","lab":"minimax","score":80.5,"best_input_per_m":0.225,"best_output_per_m":0.9,"points_per_dollar":13211.88},{"model":"alibaba/qwen3.7-max","lab":"alibaba","score":80.4,"best_input_per_m":0.825,"best_output_per_m":2.4755,"points_per_dollar":4413.1},{"model":"moonshotai/kimi-k2.6","lab":"moonshotai","score":80.2,"best_input_per_m":0.275,"best_output_per_m":1.1,"points_per_dollar":8628.29},{"model":"minimax/MiniMax-M2.7","lab":"minimax","score":79.9,"best_input_per_m":0.08,"best_output_per_m":0.32,"points_per_dollar":36739.01},{"model":"deepseek/deepseek-v4-flash","lab":"deepseek","score":79,"best_input_per_m":0.035,"best_output_per_m":0.07,"points_per_dollar":132151.22},{"model":"xiaomi/mimo-v2.5-pro","lab":"xiaomi","score":78.9,"best_input_per_m":0.4,"best_output_per_m":0.8,"points_per_dollar":12756.26},{"model":"tencent/hy3","lab":"tencent","score":78,"best_input_per_m":0.066,"best_output_per_m":0.26,"points_per_dollar":41052.63},{"model":"mistral/mistral-medium-2604","lab":"mistral","score":77.6,"best_input_per_m":1.5,"best_output_per_m":7.5,"points_per_dollar":1656},{"model":"mistral/mistral-medium-latest","lab":"mistral","score":77.6,"best_input_per_m":0.4,"best_output_per_m":2,"points_per_dollar":5000},{"model":"alibaba/qwen3.6-27b","lab":"alibaba","score":77.2,"best_input_per_m":0.2,"best_output_per_m":1.5,"points_per_dollar":8658.59},{"model":"stepfun/step-3.7-flash","lab":"stepfun","score":76.5,"best_input_per_m":0.185,"best_output_per_m":1.11,"points_per_dollar":11151.93},{"model":"alibaba/qwen3.5-397b-a17b","lab":"alibaba","score":76.4,"best_input_per_m":0.172,"best_output_per_m":1.032,"points_per_dollar":11979.13},{"model":"meta/muse-glimmer-30b","lab":"meta","score":76,"best_input_per_m":0.2,"best_output_per_m":0.8,"points_per_dollar":13818.18},{"model":"minimax/MiniMax-M2.5","lab":"minimax","score":75.8,"best_input_per_m":0.22,"best_output_per_m":0.88,"points_per_dollar":10193.65},{"model":"stepfun/step-3.5-flash","lab":"stepfun","score":74.4,"best_input_per_m":0.09,"best_output_per_m":0.3,"points_per_dollar":34540.39},{"model":"tencent/hy3-preview","lab":"tencent","score":74.4,"best_input_per_m":0.066,"best_output_per_m":0.26,"points_per_dollar":39157.89},{"model":"minimax/MiniMax-M2.1","lab":"minimax","score":74,"best_input_per_m":0.27,"best_output_per_m":1.1,"points_per_dollar":8020.81},{"model":"zhipuai/glm-4.7","lab":"zhipuai","score":73.8,"best_input_per_m":0.2,"best_output_per_m":0.8,"points_per_dollar":12466.22},{"model":"alibaba/qwen3.6-35b-a3b","lab":"alibaba","score":73.4,"best_input_per_m":0.07,"best_output_per_m":0.42,"points_per_dollar":26479.08},{"model":"zhipuai/glm-5","lab":"zhipuai","score":72.8,"best_input_per_m":0.35,"best_output_per_m":1.4,"points_per_dollar":7027.03},{"model":"alibaba/qwen3.5-27b","lab":"alibaba","score":72.4,"best_input_per_m":0.086,"best_output_per_m":0.688,"points_per_dollar":17881.49},{"model":"alibaba/qwen3.5-122b-a10b","lab":"alibaba","score":72,"best_input_per_m":0.115,"best_output_per_m":0.917,"points_per_dollar":13335.31},{"model":"microsoft/mai-code-1-flash","lab":"microsoft","score":71.6,"best_input_per_m":0.75,"best_output_per_m":4.5,"points_per_dollar":2634.29},{"model":"moonshotai/kimi-k2-thinking","lab":"moonshotai","score":71.3,"best_input_per_m":0.47,"best_output_per_m":2,"points_per_dollar":5195.88},{"model":"poolside/laguna-xs-2.1","lab":"poolside","score":70.9,"best_input_per_m":0.06,"best_output_per_m":0.12,"points_per_dollar":60289.12},{"model":"moonshotai/kimi-k2.5","lab":"moonshotai","score":70.8,"best_input_per_m":0.3,"best_output_per_m":1.9,"points_per_dollar":5718.9},{"model":"nvidia/nemotron-3-ultra-550b-a55b","lab":"nvidia","score":70.7,"best_input_per_m":0.1,"best_output_per_m":0.1,"points_per_dollar":37606.38},{"model":"minimax/MiniMax-M2","lab":"minimax","score":69.4,"best_input_per_m":0.12,"best_output_per_m":0.48,"points_per_dollar":17110.45},{"model":"cohere/north-mini-code-1-0","lab":"cohere","score":67.6,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null},{"model":"mistral/devstral-medium-2507","lab":"mistral","score":61.6,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null},{"model":"zhipuai/glm-4.7-flash","lab":"zhipuai","score":59.2,"best_input_per_m":0.06,"best_output_per_m":0.4,"points_per_dollar":25739.13},{"model":"mistral/devstral-small-2507","lab":"mistral","score":53.6,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null}]},{"name":"Terminal-Bench","slug":"terminal-bench","metric":"success rate","entries":[{"model":"openai/gpt-5.6-sol","lab":"openai","score":88.8,"best_input_per_m":2,"best_output_per_m":10,"points_per_dollar":1401.07},{"model":"moonshotai/kimi-k3","lab":"moonshotai","score":88.3,"best_input_per_m":2,"best_output_per_m":8,"points_per_dollar":1669.19},{"model":"anthropic/claude-fable-5","lab":"anthropic","score":88,"best_input_per_m":3,"best_output_per_m":18.5,"points_per_dollar":791.22},{"model":"openai/gpt-5.6-terra","lab":"openai","score":87.4,"best_input_per_m":1.5,"best_output_per_m":2,"points_per_dollar":4514.46},{"model":"alibaba/qwen3.8-max-preview","lab":"alibaba","score":86.6,"best_input_per_m":2,"best_output_per_m":6,"points_per_dollar":2018.65},{"model":"google/gemini-3.7-flash","lab":"google","score":85.8,"best_input_per_m":0.75,"best_output_per_m":3.75,"points_per_dollar":3873.87},{"model":"openai/gpt-5.6-luna","lab":"openai","score":84.7,"best_input_per_m":0.06,"best_output_per_m":0.37,"points_per_dollar":31628.08},{"model":"openai/gpt-5.5","lab":"openai","score":84.1,"best_input_per_m":0.1875,"best_output_per_m":1.125,"points_per_dollar":10240.49},{"model":"xai/grok-4.5","lab":"xai","score":83.3,"best_input_per_m":2,"best_output_per_m":6,"points_per_dollar":1922.9},{"model":"deepseek/deepseek-v4-flash-0731","lab":"deepseek","score":82.7,"best_input_per_m":0.035,"best_output_per_m":0.07,"points_per_dollar":138340.58},{"model":"zhipuai/glm-5.2","lab":"zhipuai","score":82.7,"best_input_per_m":0.3,"best_output_per_m":1.05,"points_per_dollar":11344.31},{"model":"anthropic/claude-sonnet-5","lab":"anthropic","score":80.4,"best_input_per_m":1.44,"best_output_per_m":7.2,"points_per_dollar":1439},{"model":"meta/muse-spark-1.1","lab":"meta","score":80,"best_input_per_m":1.25,"best_output_per_m":4.25,"points_per_dollar":2734.11},{"model":"google/gemini-3.6-flash","lab":"google","score":78,"best_input_per_m":0.375,"best_output_per_m":1.875,"points_per_dollar":7041.3},{"model":"google/gemini-3.5-flash","lab":"google","score":76.2,"best_input_per_m":0.1857,"best_output_per_m":1.1142,"points_per_dollar":9368.48},{"model":"anthropic/claude-opus-4-8","lab":"anthropic","score":74.6,"best_input_per_m":0.425,"best_output_per_m":2.125,"points_per_dollar":4523.95},{"model":"anthropic/claude-opus-4-7","lab":"anthropic","score":73.8,"best_input_per_m":4,"best_output_per_m":20,"points_per_dollar":1105.45},{"model":"meituan/longcat-2.0","lab":"meituan","score":70.8,"best_input_per_m":0.3,"best_output_per_m":1.2,"points_per_dollar":9230.29},{"model":"anthropic/claude-opus-4-6","lab":"anthropic","score":70.2,"best_input_per_m":4,"best_output_per_m":20,"points_per_dollar":1051.53},{"model":"openai/gpt-5.4","lab":"openai","score":69.8,"best_input_per_m":0.75,"best_output_per_m":6,"points_per_dollar":2012.69},{"model":"alibaba/qwen3.7-max","lab":"alibaba","score":69.7,"best_input_per_m":0.825,"best_output_per_m":2.4755,"points_per_dollar":3825.78},{"model":"google/gemini-3.1-pro-preview","lab":"google","score":68.3,"best_input_per_m":1,"best_output_per_m":6,"points_per_dollar":1963.91},{"model":"minimax/MiniMax-M3","lab":"minimax","score":66,"best_input_per_m":0.225,"best_output_per_m":0.9,"points_per_dollar":10832.1},{"model":"zhipuai/glm-5.1","lab":"zhipuai","score":65.1,"best_input_per_m":0.45,"best_output_per_m":2.15,"points_per_dollar":4991.57},{"model":"deepseek/deepseek-v4-pro","lab":"deepseek","score":64.7,"best_input_per_m":0.35,"best_output_per_m":0.8,"points_per_dollar":10937.92},{"model":"moonshotai/kimi-k2.6","lab":"moonshotai","score":64.3,"best_input_per_m":0.275,"best_output_per_m":1.1,"points_per_dollar":6917.7},{"model":"anthropic/claude-sonnet-4-6","lab":"anthropic","score":63.1,"best_input_per_m":0.9,"best_output_per_m":5.55,"points_per_dollar":1891.15},{"model":"openai/gpt-5.4-mini","lab":"openai","score":60,"best_input_per_m":0.375,"best_output_per_m":4,"points_per_dollar":2685.77},{"model":"stepfun/step-3.7-flash","lab":"stepfun","score":59.6,"best_input_per_m":0.185,"best_output_per_m":1.11,"points_per_dollar":8688.3},{"model":"openai/gpt-6-astra","lab":"openai","score":57.9,"best_input_per_m":10,"best_output_per_m":50,"points_per_dollar":185.34},{"model":"nvidia/nemotron-3-ultra-550b-a55b","lab":"nvidia","score":56.4,"best_input_per_m":0.1,"best_output_per_m":0.1,"points_per_dollar":30000},{"model":"microsoft/mai-code-1-flash","lab":"microsoft","score":54.8,"best_input_per_m":0.75,"best_output_per_m":4.5,"points_per_dollar":2016.19},{"model":"google/gemini-3.5-flash-lite","lab":"google","score":54,"best_input_per_m":0.15,"best_output_per_m":1.25,"points_per_dollar":7724.22},{"model":"meta/muse-glimmer-30b","lab":"meta","score":51.7,"best_input_per_m":0.2,"best_output_per_m":0.8,"points_per_dollar":9400},{"model":"minimax/MiniMax-M2.7","lab":"minimax","score":51.1,"best_input_per_m":0.08,"best_output_per_m":0.32,"points_per_dollar":23496.41},{"model":"openai/gpt-5.4-nano","lab":"openai","score":46.3,"best_input_per_m":0.18,"best_output_per_m":1.1,"points_per_dollar":6990.58},{"model":"poolside/laguna-xs-2.1","lab":"poolside","score":37.5,"best_input_per_m":0.06,"best_output_per_m":0.12,"points_per_dollar":31887.76}]},{"name":"SciCode","slug":"scicode","metric":"percent correct","entries":[{"model":"sakana/fugu","lab":"sakana","score":60.1,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null},{"model":"sakana/fugu-ultra","lab":"sakana","score":58.7,"best_input_per_m":5,"best_output_per_m":30,"points_per_dollar":198.98},{"model":"alibaba/qwen3.7-max","lab":"alibaba","score":53.5,"best_input_per_m":0.825,"best_output_per_m":2.4755,"points_per_dollar":2125.78},{"model":"google/gemini-2.5-pro","lab":"google","score":42.8,"best_input_per_m":0.87,"best_output_per_m":7,"points_per_dollar":627.2},{"model":"openai/gpt-5-codex","lab":"openai","score":40.9,"best_input_per_m":1.07,"best_output_per_m":8.5,"points_per_dollar":493.43},{"model":"stepfun/step-3.5-flash","lab":"stepfun","score":40.4,"best_input_per_m":0.09,"best_output_per_m":0.3,"points_per_dollar":13333.33},{"model":"stepfun/step-3.7-flash","lab":"stepfun","score":40,"best_input_per_m":0.185,"best_output_per_m":1.11,"points_per_dollar":3664.68},{"model":"google/gemini-2.5-flash","lab":"google","score":39.4,"best_input_per_m":0.09,"best_output_per_m":0.71,"points_per_dollar":5689.53},{"model":"stepfun/step-3.5-flash-2603","lab":"stepfun","score":38.5,"best_input_per_m":0.1,"best_output_per_m":0.3,"points_per_dollar":12622.95},{"model":"zhipuai/glm-4.6","lab":"zhipuai","score":38.4,"best_input_per_m":0.286,"best_output_per_m":1.142,"points_per_dollar":3362.23},{"model":"alibaba/qwen3-max","lab":"alibaba","score":38.3,"best_input_per_m":0.36,"best_output_per_m":1.43,"points_per_dollar":2677.39},{"model":"mistral/mistral-small-2603","lab":"mistral","score":38,"best_input_per_m":0.143,"best_output_per_m":0.568,"points_per_dollar":6687.79},{"model":"mistral/mistral-large-2512","lab":"mistral","score":36.2,"best_input_per_m":0.5,"best_output_per_m":1.5,"points_per_dollar":2373.77},{"model":"deepseek/deepseek-r1","lab":"deepseek","score":35.7,"best_input_per_m":0.4,"best_output_per_m":1.7,"points_per_dollar":2106.19},{"model":"zhipuai/glm-4.5","lab":"zhipuai","score":34.8,"best_input_per_m":0.286,"best_output_per_m":1.142,"points_per_dollar":3047.02},{"model":"openai/gpt-4o-2024-11-20","lab":"openai","score":33.3,"best_input_per_m":2.5,"best_output_per_m":10,"points_per_dollar":333},{"model":"mistral/devstral-2512","lab":"mistral","score":33.1,"best_input_per_m":0.4,"best_output_per_m":2,"points_per_dollar":1671.72},{"model":"mistral/mistral-medium-2505","lab":"mistral","score":33.1,"best_input_per_m":0.4,"best_output_per_m":2,"points_per_dollar":1671.72},{"model":"openai/gpt-4o-2024-08-06","lab":"openai","score":33.1,"best_input_per_m":2.5,"best_output_per_m":10,"points_per_dollar":331},{"model":"openai/gpt-4-turbo","lab":"openai","score":31.9,"best_input_per_m":9,"best_output_per_m":27,"points_per_dollar":116.21},{"model":"openai/gpt-4o-2024-05-13","lab":"openai","score":30.9,"best_input_per_m":5,"best_output_per_m":15,"points_per_dollar":202.62},{"model":"zhipuai/glm-4.5-air","lab":"zhipuai","score":30.6,"best_input_per_m":0.1,"best_output_per_m":0.5,"points_per_dollar":6181.82},{"model":"mistral/mistral-large-2411","lab":"mistral","score":29.2,"best_input_per_m":2.006,"best_output_per_m":6.001,"points_per_dollar":478.52},{"model":"alibaba/qwen3-coder-30b-a3b-instruct","lab":"alibaba","score":27.8,"best_input_per_m":0.06,"best_output_per_m":0.25,"points_per_dollar":11142.28},{"model":"meta/llama-3.3-70b-instruct","lab":"meta","score":26,"best_input_per_m":0.05,"best_output_per_m":0.23,"points_per_dollar":11378.56},{"model":"openai/gpt-4o-mini","lab":"openai","score":22.9,"best_input_per_m":0.075,"best_output_per_m":0.3,"points_per_dollar":7633.33},{"model":"perplexity/sonar","lab":"perplexity","score":22.9,"best_input_per_m":0.25,"best_output_per_m":2.5,"points_per_dollar":944.33},{"model":"perplexity/sonar-pro","lab":"perplexity","score":22.6,"best_input_per_m":3,"best_output_per_m":15,"points_per_dollar":152.19},{"model":"zhipuai/glm-4.5v","lab":"zhipuai","score":22.1,"best_input_per_m":0.29,"best_output_per_m":0.86,"points_per_dollar":2525.71},{"model":"google/gemini-2.5-flash-lite","lab":"google","score":19.3,"best_input_per_m":0.1,"best_output_per_m":0.1,"points_per_dollar":16782.61}]},{"name":"Aider Polyglot","slug":"aider-polyglot","metric":"percent correct","entries":[{"model":"openai/gpt-5","lab":"openai","score":88,"best_input_per_m":0.2,"best_output_per_m":1.2,"points_per_dollar":11992.37},{"model":"openai/o3-pro","lab":"openai","score":84.9,"best_input_per_m":20,"best_output_per_m":40,"points_per_dollar":178.36},{"model":"google/gemini-2.5-pro","lab":"google","score":83.1,"best_input_per_m":0.87,"best_output_per_m":7,"points_per_dollar":2055.47},{"model":"openai/o3","lab":"openai","score":81.3,"best_input_per_m":1.8,"best_output_per_m":7.2,"points_per_dollar":1642.42},{"model":"deepseek/deepseek-reasoner","lab":"deepseek","score":74.2,"best_input_per_m":0.147,"best_output_per_m":0.295,"points_per_dollar":29632.59},{"model":"anthropic/claude-opus-4-20250514","lab":"anthropic","score":72,"best_input_per_m":13.5,"best_output_per_m":67.5,"points_per_dollar":137.46},{"model":"openai/o4-mini","lab":"openai","score":72,"best_input_per_m":0.99,"best_output_per_m":4,"points_per_dollar":2623.33},{"model":"deepseek/deepseek-chat","lab":"deepseek","score":70.2,"best_input_per_m":0.147,"best_output_per_m":0.295,"points_per_dollar":28808.27},{"model":"anthropic/claude-3-7-sonnet-20250219","lab":"anthropic","score":64.9,"best_input_per_m":2.6,"best_output_per_m":13,"points_per_dollar":788.54},{"model":"openai/o1","lab":"openai","score":61.7,"best_input_per_m":14,"best_output_per_m":54,"points_per_dollar":133.2},{"model":"anthropic/claude-sonnet-4-20250514","lab":"anthropic","score":61.3,"best_input_per_m":2.7,"best_output_per_m":13.5,"points_per_dollar":585.15},{"model":"openai/o3-mini","lab":"openai","score":60.4,"best_input_per_m":0.99,"best_output_per_m":4,"points_per_dollar":1794.31},{"model":"alibaba/qwen3-235b-a22b","lab":"alibaba","score":59.6,"best_input_per_m":0.3,"best_output_per_m":0.5,"points_per_dollar":11078.07},{"model":"deepseek/deepseek-r1","lab":"deepseek","score":56.9,"best_input_per_m":0.4,"best_output_per_m":1.7,"points_per_dollar":4611.02},{"model":"google/gemini-2.5-flash","lab":"google","score":55.1,"best_input_per_m":0.09,"best_output_per_m":0.71,"points_per_dollar":11498.33},{"model":"openai/gpt-4.1","lab":"openai","score":52.4,"best_input_per_m":1.8,"best_output_per_m":7.2,"points_per_dollar":1058.59},{"model":"alibaba/qwen3-32b","lab":"alibaba","score":40,"best_input_per_m":0.08,"best_output_per_m":0.28,"points_per_dollar":15974.44},{"model":"openai/gpt-4.1-mini","lab":"openai","score":32.4,"best_input_per_m":0.36,"best_output_per_m":1.4,"points_per_dollar":3340.21},{"model":"anthropic/claude-3-5-haiku-20241022","lab":"anthropic","score":28,"best_input_per_m":0.68,"best_output_per_m":3.4,"points_per_dollar":1299.35},{"model":"openai/gpt-4o","lab":"openai","score":23.1,"best_input_per_m":1.25,"best_output_per_m":5,"points_per_dollar":624.32},{"model":"openai/gpt-4o-2024-08-06","lab":"openai","score":23.1,"best_input_per_m":2.5,"best_output_per_m":10,"points_per_dollar":312.16},{"model":"alibaba/qwen-max","lab":"alibaba","score":21.8,"best_input_per_m":0.35,"best_output_per_m":1.38,"points_per_dollar":2326.67},{"model":"openai/gpt-4o-2024-11-20","lab":"openai","score":18.2,"best_input_per_m":2.5,"best_output_per_m":10,"points_per_dollar":245.95},{"model":"meta/llama-4-maverick-17b-instruct","lab":"meta","score":15.6,"best_input_per_m":0.14,"best_output_per_m":0.59,"points_per_dollar":3195.41},{"model":"cohere/command-a-03-2025","lab":"cohere","score":12,"best_input_per_m":2.5,"best_output_per_m":10,"points_per_dollar":162.16},{"model":"mistral/codestral-latest","lab":"mistral","score":11.1,"best_input_per_m":0.3,"best_output_per_m":0.9,"points_per_dollar":1742},{"model":"openai/gpt-4.1-nano","lab":"openai","score":8.9,"best_input_per_m":0.08,"best_output_per_m":0.26,"points_per_dollar":3702.16},{"model":"openai/gpt-4o-mini","lab":"openai","score":3.6,"best_input_per_m":0.075,"best_output_per_m":0.3,"points_per_dollar":1621.62}]},{"name":"Artificial Analysis Coding Index","slug":"artificial-analysis-coding-index","metric":"index","entries":[{"model":"openai/gpt-5-codex","lab":"openai","score":38.9,"best_input_per_m":1.07,"best_output_per_m":8.5,"points_per_dollar":6889.41},{"model":"stepfun/step-3.7-flash","lab":"stepfun","score":37.1,"best_input_per_m":0.185,"best_output_per_m":1.11,"points_per_dollar":46260.79},{"model":"stepfun/step-3.5-flash-2603","lab":"stepfun","score":34.6,"best_input_per_m":0.1,"best_output_per_m":0.3,"points_per_dollar":122045.86},{"model":"cohere/north-mini-code-1-0","lab":"cohere","score":33.4,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null},{"model":"google/gemini-2.5-pro","lab":"google","score":32,"best_input_per_m":0.87,"best_output_per_m":7,"points_per_dollar":6903.47},{"model":"stepfun/step-3.5-flash","lab":"stepfun","score":31.6,"best_input_per_m":0.09,"best_output_per_m":0.3,"points_per_dollar":116712.83},{"model":"zhipuai/glm-4.6","lab":"zhipuai","score":29.5,"best_input_per_m":0.286,"best_output_per_m":1.142,"points_per_dollar":28880.51},{"model":"alibaba/qwen3-max","lab":"alibaba","score":26.4,"best_input_per_m":0.36,"best_output_per_m":1.43,"points_per_dollar":22080.96},{"model":"zhipuai/glm-4.5","lab":"zhipuai","score":26.3,"best_input_per_m":0.286,"best_output_per_m":1.142,"points_per_dollar":25747.71},{"model":"mistral/mistral-small-2603","lab":"mistral","score":24.3,"best_input_per_m":0.143,"best_output_per_m":0.568,"points_per_dollar":51644.44},{"model":"openai/gpt-4o-2024-05-13","lab":"openai","score":24.2,"best_input_per_m":5,"best_output_per_m":15,"points_per_dollar":1573.98},{"model":"zhipuai/glm-4.5-air","lab":"zhipuai","score":23.8,"best_input_per_m":0.1,"best_output_per_m":0.5,"points_per_dollar":58404.91},{"model":"mistral/devstral-2512","lab":"mistral","score":23.7,"best_input_per_m":0.4,"best_output_per_m":2,"points_per_dollar":15571.62},{"model":"mistral/mistral-large-2512","lab":"mistral","score":22.7,"best_input_per_m":0.5,"best_output_per_m":1.5,"points_per_dollar":16185.38},{"model":"google/gemini-2.5-flash","lab":"google","score":22.2,"best_input_per_m":0.09,"best_output_per_m":0.71,"points_per_dollar":44690.49},{"model":"openai/gpt-4-turbo","lab":"openai","score":21.5,"best_input_per_m":9,"best_output_per_m":27,"points_per_dollar":776.87},{"model":"alibaba/qwen3-coder-30b-a3b-instruct","lab":"alibaba","score":19.4,"best_input_per_m":0.06,"best_output_per_m":0.25,"points_per_dollar":93663.25},{"model":"openai/gpt-4o-2024-11-20","lab":"openai","score":16.7,"best_input_per_m":2.5,"best_output_per_m":10,"points_per_dollar":1950.36},{"model":"openai/gpt-4o-2024-08-06","lab":"openai","score":16.6,"best_input_per_m":2.5,"best_output_per_m":10,"points_per_dollar":1938.69},{"model":"deepseek/deepseek-r1","lab":"deepseek","score":15.9,"best_input_per_m":0.4,"best_output_per_m":1.7,"points_per_dollar":11197.18},{"model":"mistral/mistral-large-2411","lab":"mistral","score":13.8,"best_input_per_m":2.006,"best_output_per_m":6.001,"points_per_dollar":2456.32},{"model":"mistral/mistral-medium-2505","lab":"mistral","score":13.6,"best_input_per_m":0.4,"best_output_per_m":2,"points_per_dollar":8935.61},{"model":"openai/gpt-4","lab":"openai","score":13.1,"best_input_per_m":30,"best_output_per_m":60,"points_per_dollar":169.58},{"model":"zhipuai/glm-4.5v","lab":"zhipuai","score":10.9,"best_input_per_m":0.29,"best_output_per_m":0.86,"points_per_dollar":12292.08},{"model":"meta/llama-3.3-70b-instruct","lab":"meta","score":10.7,"best_input_per_m":0.05,"best_output_per_m":0.23,"points_per_dollar":57449.66},{"model":"openai/gpt-3.5-turbo","lab":"openai","score":10.7,"best_input_per_m":0.5,"best_output_per_m":1.5,"points_per_dollar":7711.71},{"model":"google/gemini-2.5-flash-lite","lab":"google","score":9.5,"best_input_per_m":0.1,"best_output_per_m":0.1,"points_per_dollar":45783.13}]},{"name":"Terminal-Bench Hard","slug":"terminal-bench-hard","metric":"success rate","entries":[{"model":"openai/gpt-5-codex","lab":"openai","score":37.9,"best_input_per_m":1.07,"best_output_per_m":8.5,"points_per_dollar":770.69},{"model":"stepfun/step-3.7-flash","lab":"stepfun","score":35.6,"best_input_per_m":0.185,"best_output_per_m":1.11,"points_per_dollar":5189.66},{"model":"stepfun/step-3.5-flash-2603","lab":"stepfun","score":32.6,"best_input_per_m":0.1,"best_output_per_m":0.3,"points_per_dollar":14764.49},{"model":"stepfun/step-3.5-flash","lab":"stepfun","score":27.3,"best_input_per_m":0.09,"best_output_per_m":0.3,"points_per_dollar":12674.09},{"model":"google/gemini-2.5-pro","lab":"google","score":26.5,"best_input_per_m":0.87,"best_output_per_m":7,"points_per_dollar":655.47},{"model":"zhipuai/glm-4.6","lab":"zhipuai","score":25,"best_input_per_m":0.286,"best_output_per_m":1.142,"points_per_dollar":2588.85},{"model":"zhipuai/glm-4.5","lab":"zhipuai","score":22,"best_input_per_m":0.286,"best_output_per_m":1.142,"points_per_dollar":2278.19},{"model":"alibaba/qwen3-max","lab":"alibaba","score":20.5,"best_input_per_m":0.36,"best_output_per_m":1.43,"points_per_dollar":2113.66},{"model":"zhipuai/glm-4.5-air","lab":"zhipuai","score":20.5,"best_input_per_m":0.1,"best_output_per_m":0.5,"points_per_dollar":5283.51},{"model":"mistral/devstral-2512","lab":"mistral","score":18.9,"best_input_per_m":0.4,"best_output_per_m":2,"points_per_dollar":1512.48},{"model":"mistral/mistral-small-2603","lab":"mistral","score":17.4,"best_input_per_m":0.143,"best_output_per_m":0.568,"points_per_dollar":4665.13},{"model":"mistral/mistral-large-2512","lab":"mistral","score":15.9,"best_input_per_m":0.5,"best_output_per_m":1.5,"points_per_dollar":1497.18},{"model":"alibaba/qwen3-coder-30b-a3b-instruct","lab":"alibaba","score":15.2,"best_input_per_m":0.06,"best_output_per_m":0.25,"points_per_dollar":8801.39},{"model":"google/gemini-2.5-flash","lab":"google","score":13.6,"best_input_per_m":0.09,"best_output_per_m":0.71,"points_per_dollar":2838.06},{"model":"openai/gpt-4o-2024-08-06","lab":"openai","score":8.3,"best_input_per_m":2.5,"best_output_per_m":10,"points_per_dollar":112.16},{"model":"openai/gpt-4o-2024-11-20","lab":"openai","score":8.3,"best_input_per_m":2.5,"best_output_per_m":10,"points_per_dollar":112.16},{"model":"deepseek/deepseek-r1","lab":"deepseek","score":6.1,"best_input_per_m":0.4,"best_output_per_m":1.7,"points_per_dollar":494.33},{"model":"mistral/mistral-large-2411","lab":"mistral","score":6.1,"best_input_per_m":2.006,"best_output_per_m":6.001,"points_per_dollar":143.47},{"model":"zhipuai/glm-4.5v","lab":"zhipuai","score":5.3,"best_input_per_m":0.29,"best_output_per_m":0.86,"points_per_dollar":638.4},{"model":"google/gemini-2.5-flash-lite","lab":"google","score":4.5,"best_input_per_m":0.1,"best_output_per_m":0.1,"points_per_dollar":2393.62},{"model":"mistral/mistral-medium-2505","lab":"mistral","score":3.8,"best_input_per_m":0.4,"best_output_per_m":2,"points_per_dollar":304.1},{"model":"meta/llama-3.3-70b-instruct","lab":"meta","score":3,"best_input_per_m":0.05,"best_output_per_m":0.23,"points_per_dollar":1840.49}]},{"name":"Humanity's Last Exam","slug":"humanity-s-last-exam","metric":"accuracy","entries":[{"model":"anthropic/claude-opus-5","lab":"anthropic","score":64.7,"best_input_per_m":5,"best_output_per_m":25,"points_per_dollar":261.41},{"model":"anthropic/claude-fable-5","lab":"anthropic","score":64.5,"best_input_per_m":3,"best_output_per_m":18.5,"points_per_dollar":354.88},{"model":"meta/muse-spark-1.1","lab":"meta","score":62.1,"best_input_per_m":1.25,"best_output_per_m":4.25,"points_per_dollar":1448.4},{"model":"openai/gpt-5.4-pro","lab":"openai","score":58.7,"best_input_per_m":27,"best_output_per_m":160,"points_per_dollar":37.29},{"model":"anthropic/claude-opus-4-8","lab":"anthropic","score":57.9,"best_input_per_m":0.425,"best_output_per_m":2.125,"points_per_dollar":2752.23},{"model":"openai/gpt-5.5-pro","lab":"openai","score":57.2,"best_input_per_m":27.2727,"best_output_per_m":163.6364,"points_per_dollar":35.55},{"model":"openai/gpt-6-astra","lab":"openai","score":57.2,"best_input_per_m":10,"best_output_per_m":50,"points_per_dollar":115.56},{"model":"alibaba/qwen3.8-max-preview","lab":"alibaba","score":56.2,"best_input_per_m":2,"best_output_per_m":6,"points_per_dollar":921.31},{"model":"anthropic/claude-opus-4-7","lab":"anthropic","score":54.7,"best_input_per_m":4,"best_output_per_m":20,"points_per_dollar":1439.47},{"model":"zhipuai/glm-5.2","lab":"zhipuai","score":54.7,"best_input_per_m":0.3,"best_output_per_m":1.05,"points_per_dollar":5172.58},{"model":"openai/gpt-5.5","lab":"openai","score":52.2,"best_input_per_m":0.1875,"best_output_per_m":1.125,"points_per_dollar":4718.64},{"model":"openai/gpt-5.4","lab":"openai","score":52.1,"best_input_per_m":0.75,"best_output_per_m":6,"points_per_dollar":890.6},{"model":"sakana/fugu-ultra","lab":"sakana","score":50,"best_input_per_m":5,"best_output_per_m":30,"points_per_dollar":169.49},{"model":"sakana/fugu","lab":"sakana","score":47.2,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null},{"model":"stepfun/step-3.7-flash","lab":"stepfun","score":47.2,"best_input_per_m":0.185,"best_output_per_m":1.11,"points_per_dollar":4324.32},{"model":"anthropic/claude-sonnet-4-6","lab":"anthropic","score":46.8,"best_input_per_m":0.9,"best_output_per_m":5.55,"points_per_dollar":858.32},{"model":"google/gemini-3.1-pro-preview","lab":"google","score":44.4,"best_input_per_m":1,"best_output_per_m":6,"points_per_dollar":752.54},{"model":"openai/gpt-5.4-mini","lab":"openai","score":41.5,"best_input_per_m":0.375,"best_output_per_m":4,"points_per_dollar":1070.97},{"model":"alibaba/qwen3.7-max","lab":"alibaba","score":41.4,"best_input_per_m":0.825,"best_output_per_m":2.4755,"points_per_dollar":1644.99},{"model":"google/gemini-3.5-flash","lab":"google","score":40.2,"best_input_per_m":0.1857,"best_output_per_m":1.1142,"points_per_dollar":3669.12},{"model":"openai/gpt-5.4-nano","lab":"openai","score":37.7,"best_input_per_m":0.18,"best_output_per_m":1.1,"points_per_dollar":3487.51},{"model":"nvidia/nemotron-3-ultra-550b-a55b","lab":"nvidia","score":37.4,"best_input_per_m":0.1,"best_output_per_m":0.1,"points_per_dollar":32521.74}]},{"name":"GPQA Diamond","slug":"gpqa-diamond","metric":"accuracy","entries":[{"model":"openai/gpt-6-astra","lab":"openai","score":96,"best_input_per_m":10,"best_output_per_m":50,"points_per_dollar":193.94},{"model":"sakana/fugu","lab":"sakana","score":95.5,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null},{"model":"sakana/fugu-ultra","lab":"sakana","score":95.5,"best_input_per_m":5,"best_output_per_m":30,"points_per_dollar":323.73},{"model":"openai/gpt-5.6-sol","lab":"openai","score":94.6,"best_input_per_m":2,"best_output_per_m":10,"points_per_dollar":955.56},{"model":"openai/gpt-5.4-pro","lab":"openai","score":94.4,"best_input_per_m":27,"best_output_per_m":160,"points_per_dollar":59.97},{"model":"google/gemini-3.1-pro-preview","lab":"google","score":94.3,"best_input_per_m":1,"best_output_per_m":6,"points_per_dollar":1598.31},{"model":"anthropic/claude-opus-4-7","lab":"anthropic","score":94.2,"best_input_per_m":4,"best_output_per_m":20,"points_per_dollar":2478.95},{"model":"openai/gpt-5.5","lab":"openai","score":93.6,"best_input_per_m":0.1875,"best_output_per_m":1.125,"points_per_dollar":8461.02},{"model":"openai/gpt-5.6-terra","lab":"openai","score":92.9,"best_input_per_m":1.5,"best_output_per_m":2,"points_per_dollar":4222.73},{"model":"openai/gpt-5.4","lab":"openai","score":92.8,"best_input_per_m":0.75,"best_output_per_m":6,"points_per_dollar":1586.32},{"model":"alibaba/qwen3.8-max-preview","lab":"alibaba","score":92.6,"best_input_per_m":2,"best_output_per_m":6,"points_per_dollar":1518.03},{"model":"alibaba/qwen3.7-max","lab":"alibaba","score":92.4,"best_input_per_m":0.825,"best_output_per_m":2.4755,"points_per_dollar":3671.44},{"model":"openai/gpt-5.6-luna","lab":"openai","score":92.3,"best_input_per_m":0.06,"best_output_per_m":0.37,"points_per_dollar":25392.02},{"model":"zhipuai/glm-5.2","lab":"zhipuai","score":91.2,"best_input_per_m":0.3,"best_output_per_m":1.05,"points_per_dollar":8624.11},{"model":"meituan/longcat-2.0","lab":"meituan","score":88.9,"best_input_per_m":0.3,"best_output_per_m":1.2,"points_per_dollar":7408.33},{"model":"openai/gpt-5.4-mini","lab":"openai","score":88,"best_input_per_m":0.375,"best_output_per_m":4,"points_per_dollar":2270.97},{"model":"xiaomi/mimo-v2.5-pro","lab":"xiaomi","score":86.6,"best_input_per_m":0.4,"best_output_per_m":0.8,"points_per_dollar":10309.52},{"model":"microsoft/mai-code-1-flash","lab":"microsoft","score":84.6,"best_input_per_m":0.75,"best_output_per_m":4.5,"points_per_dollar":1911.86},{"model":"meta/muse-glimmer-30b","lab":"meta","score":83.5,"best_input_per_m":0.2,"best_output_per_m":0.8,"points_per_dollar":10437.5},{"model":"openai/gpt-5.4-nano","lab":"openai","score":82.8,"best_input_per_m":0.18,"best_output_per_m":1.1,"points_per_dollar":7659.57}]},{"name":"OSWorld-Verified","slug":"osworld-verified","metric":"success rate","entries":[{"model":"alibaba/qwen3.8-max-preview","lab":"alibaba","score":86.1,"best_input_per_m":2,"best_output_per_m":6,"points_per_dollar":2006.99},{"model":"anthropic/claude-fable-5","lab":"anthropic","score":85,"best_input_per_m":3,"best_output_per_m":18.5,"points_per_dollar":764.25},{"model":"anthropic/claude-opus-4-8","lab":"anthropic","score":83.4,"best_input_per_m":0.425,"best_output_per_m":2.125,"points_per_dollar":5057.61},{"model":"google/gemini-3.6-flash","lab":"google","score":83,"best_input_per_m":0.375,"best_output_per_m":1.875,"points_per_dollar":7492.67},{"model":"anthropic/claude-sonnet-5","lab":"anthropic","score":81.2,"best_input_per_m":1.44,"best_output_per_m":7.2,"points_per_dollar":1453.32},{"model":"meta/muse-spark-1.1","lab":"meta","score":80.8,"best_input_per_m":1.25,"best_output_per_m":4.25,"points_per_dollar":2761.45},{"model":"openai/gpt-5.5","lab":"openai","score":78.7,"best_input_per_m":0.1875,"best_output_per_m":1.125,"points_per_dollar":9582.95},{"model":"anthropic/claude-sonnet-4-6","lab":"anthropic","score":78.5,"best_input_per_m":0.9,"best_output_per_m":5.55,"points_per_dollar":2352.69},{"model":"google/gemini-3.5-flash","lab":"google","score":78.4,"best_input_per_m":0.1857,"best_output_per_m":1.1142,"points_per_dollar":9638.96},{"model":"anthropic/claude-opus-4-7","lab":"anthropic","score":78,"best_input_per_m":4,"best_output_per_m":20,"points_per_dollar":1168.36},{"model":"google/gemini-3.1-pro-preview","lab":"google","score":76.2,"best_input_per_m":1,"best_output_per_m":6,"points_per_dollar":2191.07},{"model":"openai/gpt-5.4","lab":"openai","score":75,"best_input_per_m":0.75,"best_output_per_m":6,"points_per_dollar":2162.63},{"model":"google/gemini-3.5-flash-lite","lab":"google","score":74,"best_input_per_m":0.15,"best_output_per_m":1.25,"points_per_dollar":10585.04},{"model":"openai/gpt-5.4-mini","lab":"openai","score":72.1,"best_input_per_m":0.375,"best_output_per_m":4,"points_per_dollar":3227.39},{"model":"minimax/MiniMax-M3","lab":"minimax","score":70.06,"best_input_per_m":0.225,"best_output_per_m":0.9,"points_per_dollar":11498.44},{"model":"meta/muse-glimmer-30b","lab":"meta","score":65.9,"best_input_per_m":0.2,"best_output_per_m":0.8,"points_per_dollar":11981.82},{"model":"openai/gpt-5.4-nano","lab":"openai","score":39,"best_input_per_m":0.18,"best_output_per_m":1.1,"points_per_dollar":5888.39}]},{"name":"BrowseComp","slug":"browsecomp","metric":"accuracy","entries":[{"model":"openai/gpt-6-astra","lab":"openai","score":91.5,"best_input_per_m":10,"best_output_per_m":50,"points_per_dollar":292.89},{"model":"moonshotai/kimi-k3","lab":"moonshotai","score":91.2,"best_input_per_m":2,"best_output_per_m":8,"points_per_dollar":1724.01},{"model":"anthropic/claude-opus-5","lab":"anthropic","score":90.8,"best_input_per_m":5,"best_output_per_m":25,"points_per_dollar":573.05},{"model":"openai/gpt-5.6-sol","lab":"openai","score":90.4,"best_input_per_m":2,"best_output_per_m":10,"points_per_dollar":1426.32},{"model":"openai/gpt-5.5-pro","lab":"openai","score":90.1,"best_input_per_m":27.2727,"best_output_per_m":163.6364,"points_per_dollar":75.43},{"model":"openai/gpt-5.4-pro","lab":"openai","score":89.3,"best_input_per_m":27,"best_output_per_m":160,"points_per_dollar":76.16},{"model":"openai/gpt-5.6-terra","lab":"openai","score":87.5,"best_input_per_m":1.5,"best_output_per_m":2,"points_per_dollar":4519.63},{"model":"anthropic/claude-sonnet-5","lab":"anthropic","score":84.7,"best_input_per_m":1.44,"best_output_per_m":7.2,"points_per_dollar":1515.97},{"model":"openai/gpt-5.5","lab":"openai","score":84.4,"best_input_per_m":0.1875,"best_output_per_m":1.125,"points_per_dollar":10277.02},{"model":"minimax/MiniMax-M3","lab":"minimax","score":83.52,"best_input_per_m":0.225,"best_output_per_m":0.9,"points_per_dollar":13707.53},{"model":"openai/gpt-5.6-luna","lab":"openai","score":83.3,"best_input_per_m":0.06,"best_output_per_m":0.37,"points_per_dollar":31105.3},{"model":"openai/gpt-5.4","lab":"openai","score":82.7,"best_input_per_m":0.75,"best_output_per_m":6,"points_per_dollar":2384.66},{"model":"meituan/longcat-2.0","lab":"meituan","score":79.9,"best_input_per_m":0.3,"best_output_per_m":1.2,"points_per_dollar":10416.67},{"model":"stepfun/step-3.7-flash","lab":"stepfun","score":75.8,"best_input_per_m":0.185,"best_output_per_m":1.11,"points_per_dollar":11049.88},{"model":"nvidia/nemotron-3-ultra-550b-a55b","lab":"nvidia","score":44.4,"best_input_per_m":0.1,"best_output_per_m":0.1,"points_per_dollar":23617.02}]},{"name":"SWE-Atlas Codebase QnA","slug":"swe-atlas-codebase-qna","metric":"score","entries":[{"model":"anthropic/claude-opus-4-7","lab":"anthropic","score":81,"best_input_per_m":4,"best_output_per_m":20,"points_per_dollar":90.5},{"model":"openai/gpt-5.5","lab":"openai","score":80.8,"best_input_per_m":0.1875,"best_output_per_m":1.125,"points_per_dollar":1501.51},{"model":"zhipuai/glm-5.1","lab":"zhipuai","score":73.2,"best_input_per_m":0.45,"best_output_per_m":2.15,"points_per_dollar":822.01},{"model":"openai/gpt-5.4","lab":"openai","score":72.9,"best_input_per_m":0.75,"best_output_per_m":6,"points_per_dollar":434.9},{"model":"anthropic/claude-opus-4-6","lab":"anthropic","score":71.9,"best_input_per_m":4,"best_output_per_m":20,"points_per_dollar":80.34},{"model":"anthropic/claude-sonnet-4-6","lab":"anthropic","score":70.3,"best_input_per_m":0.9,"best_output_per_m":5.55,"points_per_dollar":355.32},{"model":"deepseek/deepseek-v4-pro","lab":"deepseek","score":67.8,"best_input_per_m":0.35,"best_output_per_m":0.8,"points_per_dollar":943.96},{"model":"moonshotai/kimi-k2.6","lab":"moonshotai","score":59.8,"best_input_per_m":0.275,"best_output_per_m":1.1,"points_per_dollar":768.39},{"model":"google/gemini-3.1-pro-preview","lab":"google","score":45.6,"best_input_per_m":1,"best_output_per_m":6,"points_per_dollar":228.93},{"model":"openai/gpt-5.3-codex","lab":"openai","score":32.6,"best_input_per_m":1.6,"best_output_per_m":13,"points_per_dollar":91.06},{"model":"zhipuai/glm-5","lab":"zhipuai","score":20.5,"best_input_per_m":0.35,"best_output_per_m":1.4,"points_per_dollar":238.58},{"model":"moonshotai/kimi-k2.5","lab":"moonshotai","score":13.1,"best_input_per_m":0.3,"best_output_per_m":1.9,"points_per_dollar":174.55},{"model":"minimax/MiniMax-M2.5","lab":"minimax","score":10.3,"best_input_per_m":0.22,"best_output_per_m":0.88,"points_per_dollar":165.44},{"model":"google/gemini-3-flash-preview","lab":"google","score":8.2,"best_input_per_m":0.07,"best_output_per_m":0.43,"points_per_dollar":407.76}]},{"name":"DeepSWE","slug":"deepswe","metric":"resolve rate","entries":[{"model":"openai/gpt-6-astra","lab":"openai","score":74.1,"best_input_per_m":10,"best_output_per_m":50,"points_per_dollar":237.2},{"model":"openai/gpt-5.6-sol","lab":"openai","score":72.7,"best_input_per_m":2,"best_output_per_m":10,"points_per_dollar":1147.05},{"model":"openai/gpt-5.6-terra","lab":"openai","score":69.6,"best_input_per_m":1.5,"best_output_per_m":2,"points_per_dollar":3595.04},{"model":"anthropic/claude-opus-5","lab":"anthropic","score":68.8,"best_input_per_m":5,"best_output_per_m":25,"points_per_dollar":434.21},{"model":"moonshotai/kimi-k3","lab":"moonshotai","score":67.5,"best_input_per_m":2,"best_output_per_m":8,"points_per_dollar":1275.99},{"model":"openai/gpt-5.6-luna","lab":"openai","score":67.2,"best_input_per_m":0.06,"best_output_per_m":0.37,"points_per_dollar":25093.35},{"model":"google/gemini-3.7-flash","lab":"google","score":65.3,"best_input_per_m":0.75,"best_output_per_m":3.75,"points_per_dollar":2948.29},{"model":"xai/grok-4.5","lab":"xai","score":62,"best_input_per_m":2,"best_output_per_m":6,"points_per_dollar":1431.21},{"model":"alibaba/qwen3.8-max-preview","lab":"alibaba","score":56.6,"best_input_per_m":2,"best_output_per_m":6,"points_per_dollar":1319.35},{"model":"deepseek/deepseek-v4-flash-0731","lab":"deepseek","score":54.4,"best_input_per_m":0.035,"best_output_per_m":0.07,"points_per_dollar":91000.33},{"model":"meta/muse-spark-1.1","lab":"meta","score":53.3,"best_input_per_m":1.25,"best_output_per_m":4.25,"points_per_dollar":1821.6},{"model":"google/gemini-3.6-flash","lab":"google","score":49,"best_input_per_m":0.375,"best_output_per_m":1.875,"points_per_dollar":4423.38},{"model":"zhipuai/glm-5.2","lab":"zhipuai","score":46.2,"best_input_per_m":0.3,"best_output_per_m":1.05,"points_per_dollar":6337.45}]},{"name":"Artificial Analysis Coding Agent Index","slug":"artificial-analysis-coding-agent-index","metric":"average pass@1","entries":[{"model":"openai/gpt-5.6-sol","lab":"openai","score":80,"best_input_per_m":2,"best_output_per_m":10,"points_per_dollar":1262.23},{"model":"openai/gpt-5.6-terra","lab":"openai","score":77.4,"best_input_per_m":1.5,"best_output_per_m":2,"points_per_dollar":3997.93},{"model":"openai/gpt-5.6-luna","lab":"openai","score":74.6,"best_input_per_m":0.06,"best_output_per_m":0.37,"points_per_dollar":27856.61},{"model":"openai/gpt-6-astra","lab":"openai","score":67,"best_input_per_m":10,"best_output_per_m":50,"points_per_dollar":214.47},{"model":"anthropic/claude-opus-4-7","lab":"anthropic","score":66.6,"best_input_per_m":4,"best_output_per_m":20,"points_per_dollar":997.6},{"model":"openai/gpt-5.5","lab":"openai","score":65.3,"best_input_per_m":0.1875,"best_output_per_m":1.125,"points_per_dollar":7951.29},{"model":"openai/gpt-5.4","lab":"openai","score":53.6,"best_input_per_m":0.75,"best_output_per_m":6,"points_per_dollar":1545.56},{"model":"zhipuai/glm-5.1","lab":"zhipuai","score":52.7,"best_input_per_m":0.45,"best_output_per_m":2.15,"points_per_dollar":4040.79},{"model":"anthropic/claude-opus-4-6","lab":"anthropic","score":51.3,"best_input_per_m":4,"best_output_per_m":20,"points_per_dollar":768.42},{"model":"moonshotai/kimi-k2.6","lab":"moonshotai","score":50.5,"best_input_per_m":0.275,"best_output_per_m":1.1,"points_per_dollar":5433.03},{"model":"deepseek/deepseek-v4-pro","lab":"deepseek","score":50.1,"best_input_per_m":0.35,"best_output_per_m":0.8,"points_per_dollar":8469.71},{"model":"anthropic/claude-sonnet-4-6","lab":"anthropic","score":49.4,"best_input_per_m":0.9,"best_output_per_m":5.55,"points_per_dollar":1480.55},{"model":"google/gemini-3.1-pro-preview","lab":"google","score":43,"best_input_per_m":1,"best_output_per_m":6,"points_per_dollar":1236.43}]},{"name":"MCP Atlas","slug":"mcp-atlas","metric":"success rate","entries":[{"model":"meta/muse-spark-1.1","lab":"meta","score":88.1,"best_input_per_m":1.25,"best_output_per_m":4.25,"points_per_dollar":3010.94},{"model":"google/gemini-3.5-flash","lab":"google","score":83.6,"best_input_per_m":0.1857,"best_output_per_m":1.1142,"points_per_dollar":10278.28},{"model":"google/gemini-3.1-pro-preview","lab":"google","score":78.2,"best_input_per_m":1,"best_output_per_m":6,"points_per_dollar":2248.58},{"model":"zhipuai/glm-5.2","lab":"zhipuai","score":76.8,"best_input_per_m":0.3,"best_output_per_m":1.05,"points_per_dollar":10534.98},{"model":"alibaba/qwen3.7-max","lab":"alibaba","score":76.4,"best_input_per_m":0.825,"best_output_per_m":2.4755,"points_per_dollar":4193.54},{"model":"moonshotai/kimi-k2.7-code","lab":"moonshotai","score":76,"best_input_per_m":0.275,"best_output_per_m":1.1,"points_per_dollar":8176.44},{"model":"meta/muse-glimmer-30b","lab":"meta","score":75.5,"best_input_per_m":0.2,"best_output_per_m":0.8,"points_per_dollar":13727.27},{"model":"openai/gpt-5.5","lab":"openai","score":75.3,"best_input_per_m":0.1875,"best_output_per_m":1.125,"points_per_dollar":9168.95},{"model":"minimax/MiniMax-M3","lab":"minimax","score":74.2,"best_input_per_m":0.225,"best_output_per_m":0.9,"points_per_dollar":12177.91},{"model":"openai/gpt-5.4-mini","lab":"openai","score":57.7,"best_input_per_m":0.375,"best_output_per_m":4,"points_per_dollar":2582.81},{"model":"openai/gpt-5.4-nano","lab":"openai","score":56.1,"best_input_per_m":0.18,"best_output_per_m":1.1,"points_per_dollar":8470.23}]},{"name":"SWE-Atlas Refactoring","slug":"swe-atlas-refactoring","metric":"score","entries":[{"model":"anthropic/claude-opus-4-7","lab":"anthropic","score":48.57,"best_input_per_m":4,"best_output_per_m":20,"points_per_dollar":54.27},{"model":"openai/gpt-5.5","lab":"openai","score":44.79,"best_input_per_m":0.1875,"best_output_per_m":1.125,"points_per_dollar":832.33},{"model":"openai/gpt-5.4","lab":"openai","score":44.29,"best_input_per_m":0.75,"best_output_per_m":6,"points_per_dollar":264.22},{"model":"openai/gpt-5.3-codex","lab":"openai","score":42.38,"best_input_per_m":1.6,"best_output_per_m":13,"points_per_dollar":118.38},{"model":"anthropic/claude-opus-4-6","lab":"anthropic","score":35.58,"best_input_per_m":4,"best_output_per_m":20,"points_per_dollar":39.75},{"model":"google/gemini-3.1-pro-preview","lab":"google","score":33.81,"best_input_per_m":1,"best_output_per_m":6,"points_per_dollar":169.74},{"model":"anthropic/claude-sonnet-4-6","lab":"anthropic","score":32.21,"best_input_per_m":0.9,"best_output_per_m":5.55,"points_per_dollar":162.8},{"model":"zhipuai/glm-5","lab":"zhipuai","score":24.24,"best_input_per_m":0.35,"best_output_per_m":1.4,"points_per_dollar":282.11},{"model":"moonshotai/kimi-k2.5","lab":"moonshotai","score":20.95,"best_input_per_m":0.3,"best_output_per_m":1.9,"points_per_dollar":279.15},{"model":"minimax/MiniMax-M2.5","lab":"minimax","score":19.52,"best_input_per_m":0.22,"best_output_per_m":0.88,"points_per_dollar":313.52},{"model":"google/gemini-3-flash-preview","lab":"google","score":10,"best_input_per_m":0.07,"best_output_per_m":0.43,"points_per_dollar":497.27}]},{"name":"MMMU Pro","slug":"mmmu-pro","metric":"accuracy","entries":[{"model":"google/gemini-3.5-flash","lab":"google","score":83.6,"best_input_per_m":0.1857,"best_output_per_m":1.1142,"points_per_dollar":7630.31},{"model":"openai/gpt-5.6-sol","lab":"openai","score":83,"best_input_per_m":2,"best_output_per_m":10,"points_per_dollar":838.38},{"model":"alibaba/qwen3.8-max-preview","lab":"alibaba","score":82.3,"best_input_per_m":2,"best_output_per_m":6,"points_per_dollar":1349.18},{"model":"openai/gpt-5.4","lab":"openai","score":81.2,"best_input_per_m":0.75,"best_output_per_m":6,"points_per_dollar":1388.03},{"model":"openai/gpt-5.5","lab":"openai","score":81.2,"best_input_per_m":0.1875,"best_output_per_m":1.125,"points_per_dollar":7340.11},{"model":"openai/gpt-5.6-terra","lab":"openai","score":80.7,"best_input_per_m":1.5,"best_output_per_m":2,"points_per_dollar":3668.18},{"model":"google/gemini-3.1-pro-preview","lab":"google","score":80.5,"best_input_per_m":1,"best_output_per_m":6,"points_per_dollar":1364.41},{"model":"openai/gpt-5.6-luna","lab":"openai","score":78.4,"best_input_per_m":0.06,"best_output_per_m":0.37,"points_per_dollar":21568.09},{"model":"openai/gpt-5.4-mini","lab":"openai","score":78,"best_input_per_m":0.375,"best_output_per_m":4,"points_per_dollar":2012.9},{"model":"openai/gpt-5.4-nano","lab":"openai","score":69.5,"best_input_per_m":0.18,"best_output_per_m":1.1,"points_per_dollar":6429.23}]},{"name":"SWE-Atlas Test Writing","slug":"swe-atlas-test-writing","metric":"score","entries":[{"model":"openai/gpt-5.4","lab":"openai","score":44.36,"best_input_per_m":0.75,"best_output_per_m":6,"points_per_dollar":264.64},{"model":"openai/gpt-5.5","lab":"openai","score":42.59,"best_input_per_m":0.1875,"best_output_per_m":1.125,"points_per_dollar":791.45},{"model":"openai/gpt-5.3-codex","lab":"openai","score":38.98,"best_input_per_m":1.6,"best_output_per_m":13,"points_per_dollar":108.88},{"model":"anthropic/claude-opus-4-6","lab":"anthropic","score":36.67,"best_input_per_m":4,"best_output_per_m":20,"points_per_dollar":40.97},{"model":"anthropic/claude-sonnet-4-6","lab":"anthropic","score":31.76,"best_input_per_m":0.9,"best_output_per_m":5.55,"points_per_dollar":160.53},{"model":"google/gemini-3-flash-preview","lab":"google","score":30.3,"best_input_per_m":0.07,"best_output_per_m":0.43,"points_per_dollar":1506.71},{"model":"google/gemini-3.1-pro-preview","lab":"google","score":29.84,"best_input_per_m":1,"best_output_per_m":6,"points_per_dollar":149.81},{"model":"zhipuai/glm-5","lab":"zhipuai","score":28.74,"best_input_per_m":0.35,"best_output_per_m":1.4,"points_per_dollar":334.48},{"model":"moonshotai/kimi-k2.5","lab":"moonshotai","score":25.77,"best_input_per_m":0.3,"best_output_per_m":1.9,"points_per_dollar":343.37},{"model":"minimax/MiniMax-M2.5","lab":"minimax","score":18.6,"best_input_per_m":0.22,"best_output_per_m":0.88,"points_per_dollar":298.75}]},{"name":"GDPval-AA","slug":"gdpval-aa","metric":"Elo","entries":[{"model":"anthropic/claude-fable-5","lab":"anthropic","score":1932,"best_input_per_m":3,"best_output_per_m":18.5,"points_per_dollar":146752.75},{"model":"anthropic/claude-opus-5","lab":"anthropic","score":1861,"best_input_per_m":5,"best_output_per_m":25,"points_per_dollar":97339},{"model":"moonshotai/kimi-k3","lab":"moonshotai","score":1668,"best_input_per_m":2,"best_output_per_m":8,"points_per_dollar":251773.58},{"model":"google/gemini-3.5-flash","lab":"google","score":1656,"best_input_per_m":0.1857,"best_output_per_m":1.1142,"points_per_dollar":1949204.16},{"model":"xai/grok-4.3","lab":"xai","score":1500,"best_input_per_m":1.25,"best_output_per_m":2.5,"points_per_dollar":533807.83},{"model":"google/gemini-3.6-flash","lab":"google","score":1421,"best_input_per_m":0.375,"best_output_per_m":1.875,"points_per_dollar":1014773.51},{"model":"google/gemini-3.1-pro-preview","lab":"google","score":1314,"best_input_per_m":1,"best_output_per_m":6,"points_per_dollar":309609.01},{"model":"google/gemini-3.5-flash-lite","lab":"google","score":1140,"best_input_per_m":0.15,"best_output_per_m":1.25,"points_per_dollar":1402861.06},{"model":"cohere/north-mini-code-1-0","lab":"cohere","score":14,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null}]},{"name":"CharXiv Reasoning","slug":"charxiv-reasoning","metric":"accuracy","entries":[{"model":"moonshotai/kimi-k3","lab":"moonshotai","score":91.3,"best_input_per_m":2,"best_output_per_m":8,"points_per_dollar":1141.25},{"model":"google/gemini-3.6-flash","lab":"google","score":89.4,"best_input_per_m":0.375,"best_output_per_m":1.875,"points_per_dollar":4816.16},{"model":"meta/muse-spark-1.1","lab":"meta","score":88.4,"best_input_per_m":1.25,"best_output_per_m":4.25,"points_per_dollar":2061.81},{"model":"sakana/fugu-ultra","lab":"sakana","score":86.6,"best_input_per_m":5,"best_output_per_m":30,"points_per_dollar":293.56},{"model":"sakana/fugu","lab":"sakana","score":85.1,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null},{"model":"google/gemini-3.5-flash","lab":"google","score":84.2,"best_input_per_m":0.1857,"best_output_per_m":1.1142,"points_per_dollar":7685.08},{"model":"google/gemini-3.1-pro-preview","lab":"google","score":83.3,"best_input_per_m":1,"best_output_per_m":6,"points_per_dollar":1411.86},{"model":"meta/muse-glimmer-30b","lab":"meta","score":78.8,"best_input_per_m":0.2,"best_output_per_m":0.8,"points_per_dollar":9850},{"model":"google/gemini-3.5-flash-lite","lab":"google","score":76.5,"best_input_per_m":0.15,"best_output_per_m":1.25,"points_per_dollar":6283.37}]},{"name":"Toolathlon","slug":"toolathlon","metric":"success rate","entries":[{"model":"openai/gpt-5.6-sol","lab":"openai","score":58,"best_input_per_m":2,"best_output_per_m":10,"points_per_dollar":915.12},{"model":"google/gemini-3.5-flash","lab":"google","score":56.5,"best_input_per_m":0.1857,"best_output_per_m":1.1142,"points_per_dollar":6946.44},{"model":"openai/gpt-5.5","lab":"openai","score":55.6,"best_input_per_m":0.1875,"best_output_per_m":1.125,"points_per_dollar":6770.17},{"model":"openai/gpt-5.6-luna","lab":"openai","score":53.4,"best_input_per_m":0.06,"best_output_per_m":0.37,"points_per_dollar":19940.25},{"model":"openai/gpt-5.6-terra","lab":"openai","score":53.1,"best_input_per_m":1.5,"best_output_per_m":2,"points_per_dollar":2742.77},{"model":"stepfun/step-3.7-flash","lab":"stepfun","score":49.5,"best_input_per_m":0.185,"best_output_per_m":1.11,"points_per_dollar":7215.95},{"model":"openai/gpt-5.4-mini","lab":"openai","score":42.9,"best_input_per_m":0.375,"best_output_per_m":4,"points_per_dollar":1920.32},{"model":"openai/gpt-5.4-nano","lab":"openai","score":35.5,"best_input_per_m":0.18,"best_output_per_m":1.1,"points_per_dollar":5359.95}]},{"name":"FrontierMath","slug":"frontiermath","metric":"accuracy","entries":[{"model":"openai/gpt-6-astra","lab":"openai","score":97.6,"best_input_per_m":10,"best_output_per_m":50,"points_per_dollar":197.17},{"model":"openai/gpt-5.6-sol","lab":"openai","score":89,"best_input_per_m":2,"best_output_per_m":10,"points_per_dollar":898.99},{"model":"openai/gpt-5.6-terra","lab":"openai","score":84.9,"best_input_per_m":1.5,"best_output_per_m":2,"points_per_dollar":3859.09},{"model":"openai/gpt-5.6-luna","lab":"openai","score":78.6,"best_input_per_m":0.06,"best_output_per_m":0.37,"points_per_dollar":21623.11},{"model":"openai/gpt-5.5-pro","lab":"openai","score":52.4,"best_input_per_m":27.2727,"best_output_per_m":163.6364,"points_per_dollar":32.56},{"model":"openai/gpt-5.5","lab":"openai","score":51.7,"best_input_per_m":0.1875,"best_output_per_m":1.125,"points_per_dollar":4673.45},{"model":"openai/gpt-5.4-pro","lab":"openai","score":50,"best_input_per_m":27,"best_output_per_m":160,"points_per_dollar":31.77},{"model":"openai/gpt-5.4","lab":"openai","score":47.6,"best_input_per_m":0.75,"best_output_per_m":6,"points_per_dollar":813.68}]},{"name":"SWE-Bench Multilingual","slug":"swe-bench-multilingual","metric":"resolve rate","entries":[{"model":"anthropic/claude-opus-5","lab":"anthropic","score":89.5,"best_input_per_m":5,"best_output_per_m":25,"points_per_dollar":564.85},{"model":"alibaba/qwen3.7-max","lab":"alibaba","score":78.3,"best_input_per_m":0.825,"best_output_per_m":2.4755,"points_per_dollar":4297.83},{"model":"anthropic/claude-sonnet-5","lab":"anthropic","score":78.3,"best_input_per_m":1.44,"best_output_per_m":7.2,"points_per_dollar":1401.42},{"model":"xai/grok-4.5","lab":"xai","score":78,"best_input_per_m":2,"best_output_per_m":6,"points_per_dollar":1800.55},{"model":"meituan/longcat-2.0","lab":"meituan","score":77.3,"best_input_per_m":0.3,"best_output_per_m":1.2,"points_per_dollar":10077.7},{"model":"nvidia/nemotron-3-ultra-550b-a55b","lab":"nvidia","score":67.7,"best_input_per_m":0.1,"best_output_per_m":0.1,"points_per_dollar":36010.64},{"model":"poolside/laguna-xs-2.1","lab":"poolside","score":63.1,"best_input_per_m":0.06,"best_output_per_m":0.12,"points_per_dollar":53656.46}]},{"name":"AutomationBench","slug":"automationbench","metric":"pass@1","entries":[{"model":"openai/gpt-6-astra","lab":"openai","score":41.4,"best_input_per_m":10,"best_output_per_m":50,"points_per_dollar":132.52},{"model":"moonshotai/kimi-k3","lab":"moonshotai","score":30.8,"best_input_per_m":2,"best_output_per_m":8,"points_per_dollar":582.23},{"model":"google/gemini-3.7-flash","lab":"google","score":30.4,"best_input_per_m":0.75,"best_output_per_m":3.75,"points_per_dollar":1372.56},{"model":"alibaba/qwen3.8-max-preview","lab":"alibaba","score":27.3,"best_input_per_m":2,"best_output_per_m":6,"points_per_dollar":636.36},{"model":"anthropic/claude-opus-5","lab":"anthropic","score":26,"best_input_per_m":5,"best_output_per_m":25,"points_per_dollar":164.09},{"model":"deepseek/deepseek-v4-flash-0731","lab":"deepseek","score":25.1,"best_input_per_m":0.035,"best_output_per_m":0.07,"points_per_dollar":41987.29},{"model":"anthropic/claude-fable-5","lab":"anthropic","score":17.4,"best_input_per_m":3,"best_output_per_m":18.5,"points_per_dollar":156.45}]},{"name":"FrontierCode","slug":"frontiercode","metric":"pass rate","entries":[{"model":"openai/gpt-6-astra","lab":"openai","score":64.5,"best_input_per_m":10,"best_output_per_m":50,"points_per_dollar":130.3},{"model":"anthropic/claude-opus-5","lab":"anthropic","score":53.4,"best_input_per_m":5,"best_output_per_m":25,"points_per_dollar":215.76},{"model":"google/gemini-3.7-flash","lab":"google","score":43.6,"best_input_per_m":0.75,"best_output_per_m":3.75,"points_per_dollar":1174.41},{"model":"anthropic/claude-sonnet-5","lab":"anthropic","score":38.8,"best_input_per_m":1.44,"best_output_per_m":7.2,"points_per_dollar":544.33},{"model":"anthropic/claude-fable-5","lab":"anthropic","score":29.3,"best_input_per_m":3,"best_output_per_m":18.5,"points_per_dollar":161.21},{"model":"anthropic/claude-opus-4-8","lab":"anthropic","score":13.4,"best_input_per_m":0.425,"best_output_per_m":2.125,"points_per_dollar":636.96}]},{"name":"ARC-AGI-2","slug":"arc-agi-2","metric":"accuracy","entries":[{"model":"anthropic/claude-opus-5","lab":"anthropic","score":90.4,"best_input_per_m":5,"best_output_per_m":25,"points_per_dollar":365.25},{"model":"openai/gpt-5.5","lab":"openai","score":85,"best_input_per_m":0.1875,"best_output_per_m":1.125,"points_per_dollar":7683.62},{"model":"openai/gpt-5.4-pro","lab":"openai","score":83.3,"best_input_per_m":27,"best_output_per_m":160,"points_per_dollar":52.92},{"model":"google/gemini-3.1-pro-preview","lab":"google","score":77.1,"best_input_per_m":1,"best_output_per_m":6,"points_per_dollar":1306.78},{"model":"openai/gpt-5.4","lab":"openai","score":73.3,"best_input_per_m":0.75,"best_output_per_m":6,"points_per_dollar":1252.99},{"model":"google/gemini-3.5-flash","lab":"google","score":72.1,"best_input_per_m":0.1857,"best_output_per_m":1.1142,"points_per_dollar":6580.69}]},{"name":"Artificial Analysis Intelligence Index","slug":"artificial-analysis-intelligence-index","metric":"index score","entries":[{"model":"openai/gpt-6-astra","lab":"openai","score":61,"best_input_per_m":10,"best_output_per_m":50,"points_per_dollar":1603.15},{"model":"openai/gpt-5.6-sol","lab":"openai","score":58.9,"best_input_per_m":2,"best_output_per_m":10,"points_per_dollar":7701.86},{"model":"openai/gpt-5.6-terra","lab":"openai","score":55,"best_input_per_m":1.5,"best_output_per_m":2,"points_per_dollar":18596.79},{"model":"xai/grok-4.3","lab":"xai","score":53,"best_input_per_m":1.25,"best_output_per_m":2.5,"points_per_dollar":18861.21},{"model":"openai/gpt-5.6-luna","lab":"openai","score":51.2,"best_input_per_m":0.06,"best_output_per_m":0.37,"points_per_dollar":183184.26},{"model":"cohere/north-mini-code-1-0","lab":"cohere","score":27.6,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null}]},{"name":"GDPval","slug":"gdpval","metric":"wins or ties","entries":[{"model":"openai/gpt-5.5","lab":"openai","score":84.9,"best_input_per_m":0.1875,"best_output_per_m":1.125,"points_per_dollar":98972.68},{"model":"openai/gpt-5.4","lab":"openai","score":83,"best_input_per_m":0.75,"best_output_per_m":6,"points_per_dollar":20860.82},{"model":"openai/gpt-5.5-pro","lab":"openai","score":82.3,"best_input_per_m":27.2727,"best_output_per_m":163.6364,"points_per_dollar":659.6},{"model":"openai/gpt-5.4-pro","lab":"openai","score":82,"best_input_per_m":27,"best_output_per_m":160,"points_per_dollar":669.25},{"model":"nvidia/nemotron-3-ultra-550b-a55b","lab":"nvidia","score":46.7,"best_input_per_m":0.1,"best_output_per_m":0.1,"points_per_dollar":225060.24},{"model":"stepfun/step-3.7-flash","lab":"stepfun","score":45.8,"best_input_per_m":0.185,"best_output_per_m":1.11,"points_per_dollar":57109.01}]},{"name":"OSWorld","slug":"osworld","metric":"success rate","entries":[{"model":"openai/gpt-6-astra","lab":"openai","score":72.6,"best_input_per_m":10,"best_output_per_m":50,"points_per_dollar":1908.02},{"model":"anthropic/claude-opus-5","lab":"anthropic","score":70.6,"best_input_per_m":5,"best_output_per_m":25,"points_per_dollar":3692.71},{"model":"openai/gpt-5.6-sol","lab":"openai","score":62.6,"best_input_per_m":2,"best_output_per_m":10,"points_per_dollar":8185.68},{"model":"openai/gpt-5.6-terra","lab":"openai","score":50.2,"best_input_per_m":1.5,"best_output_per_m":2,"points_per_dollar":16973.8},{"model":"openai/gpt-5.6-luna","lab":"openai","score":45.6,"best_input_per_m":0.06,"best_output_per_m":0.37,"points_per_dollar":163148.48}]},{"name":"τ²-Bench Telecom","slug":"bench-telecom","metric":"success rate","entries":[{"model":"openai/gpt-5.5","lab":"openai","score":98,"best_input_per_m":0.1875,"best_output_per_m":1.125,"points_per_dollar":114244.08},{"model":"xai/grok-4.3","lab":"xai","score":98,"best_input_per_m":1.25,"best_output_per_m":2.5,"points_per_dollar":34875.44},{"model":"openai/gpt-5.4-mini","lab":"openai","score":93.4,"best_input_per_m":0.375,"best_output_per_m":4,"points_per_dollar":37519.46},{"model":"openai/gpt-5.4-nano","lab":"openai","score":92.5,"best_input_per_m":0.18,"best_output_per_m":1.1,"points_per_dollar":117849.41},{"model":"cohere/north-mini-code-1-0","lab":"cohere","score":37,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null}]},{"name":"Agents' Last Exam","slug":"agents-last-exam","metric":"score","entries":[{"model":"openai/gpt-6-astra","lab":"openai","score":59.3,"best_input_per_m":10,"best_output_per_m":50,"points_per_dollar":1558.48},{"model":"openai/gpt-5.6-sol","lab":"openai","score":52.7,"best_input_per_m":2,"best_output_per_m":10,"points_per_dollar":6891.14},{"model":"openai/gpt-5.6-terra","lab":"openai","score":50.4,"best_input_per_m":1.5,"best_output_per_m":2,"points_per_dollar":17041.42},{"model":"openai/gpt-5.6-luna","lab":"openai","score":50.3,"best_input_per_m":0.06,"best_output_per_m":0.37,"points_per_dollar":179964.22},{"model":"deepseek/deepseek-v4-flash-0731","lab":"deepseek","score":25.2,"best_input_per_m":0.035,"best_output_per_m":0.07,"points_per_dollar":308351.18}]},{"name":"NL2Repo","slug":"nl2repo","metric":null,"entries":[{"model":"alibaba/qwen3.8-max-preview","lab":"alibaba","score":55.9,"best_input_per_m":2,"best_output_per_m":6,"points_per_dollar":1303.03},{"model":"deepseek/deepseek-v4-flash-0731","lab":"deepseek","score":54.2,"best_input_per_m":0.035,"best_output_per_m":0.07,"points_per_dollar":90665.77},{"model":"zhipuai/glm-5.2","lab":"zhipuai","score":48.9,"best_input_per_m":0.3,"best_output_per_m":1.05,"points_per_dollar":6707.82},{"model":"alibaba/qwen3.7-max","lab":"alibaba","score":47.2,"best_input_per_m":0.825,"best_output_per_m":2.4755,"points_per_dollar":2590.77}]},{"name":"FrontierSWE","slug":"frontierswe","metric":"dominance score","entries":[{"model":"moonshotai/kimi-k3","lab":"moonshotai","score":81.2,"best_input_per_m":2,"best_output_per_m":8,"points_per_dollar":12256.6},{"model":"zhipuai/glm-5.2","lab":"zhipuai","score":74.4,"best_input_per_m":0.3,"best_output_per_m":1.05,"points_per_dollar":80650.41},{"model":"alibaba/qwen3.8-max-preview","lab":"alibaba","score":73.5,"best_input_per_m":2,"best_output_per_m":6,"points_per_dollar":13066.67}]},{"name":"Toolathlon Verified","slug":"toolathlon-verified","metric":"pass@1","entries":[{"model":"meta/muse-spark-1.1","lab":"meta","score":75.6,"best_input_per_m":1.25,"best_output_per_m":4.25,"points_per_dollar":20086.35},{"model":"alibaba/qwen3.8-max-preview","lab":"alibaba","score":72.5,"best_input_per_m":2,"best_output_per_m":6,"points_per_dollar":12888.89},{"model":"deepseek/deepseek-v4-flash-0731","lab":"deepseek","score":70.3,"best_input_per_m":0.035,"best_output_per_m":0.07,"points_per_dollar":860201.9}]},{"name":"IFBench","slug":"ifbench","metric":"score","entries":[{"model":"alibaba/qwen3.8-max-preview","lab":"alibaba","score":82.8,"best_input_per_m":2,"best_output_per_m":6,"points_per_dollar":14720},{"model":"nvidia/nemotron-3-ultra-550b-a55b","lab":"nvidia","score":81.7,"best_input_per_m":0.1,"best_output_per_m":0.1,"points_per_dollar":393734.94},{"model":"xai/grok-4.3","lab":"xai","score":81,"best_input_per_m":1.25,"best_output_per_m":2.5,"points_per_dollar":28825.62}]},{"name":"GDM-MRCR","slug":"gdm-mrcr","metric":"accuracy","entries":[{"model":"google/gemini-3.7-flash","lab":"google","score":97,"best_input_per_m":0.75,"best_output_per_m":3.75,"points_per_dollar":667.47},{"model":"google/gemini-3.6-flash","lab":"google","score":91.8,"best_input_per_m":0.375,"best_output_per_m":1.875,"points_per_dollar":1262.58},{"model":"google/gemini-3.5-flash-lite","lab":"google","score":72.2,"best_input_per_m":0.15,"best_output_per_m":1.25,"points_per_dollar":2335.31}]},{"name":"Program Bench","slug":"program-bench","metric":null,"entries":[{"model":"moonshotai/kimi-k3","lab":"moonshotai","score":77.8,"best_input_per_m":2,"best_output_per_m":8,"points_per_dollar":11743.4},{"model":"zhipuai/glm-5.2","lab":"zhipuai","score":63.7,"best_input_per_m":0.3,"best_output_per_m":1.05,"points_per_dollar":69051.49},{"model":"moonshotai/kimi-k2.7-code","lab":"moonshotai","score":53.6,"best_input_per_m":0.275,"best_output_per_m":1.1,"points_per_dollar":54520.03}]},{"name":"SWE Marathon","slug":"swe-marathon","metric":"resolve rate","entries":[{"model":"moonshotai/kimi-k3","lab":"moonshotai","score":42,"best_input_per_m":2,"best_output_per_m":8,"points_per_dollar":6339.62},{"model":"xai/grok-4.5","lab":"xai","score":29,"best_input_per_m":2,"best_output_per_m":6,"points_per_dollar":5141.84},{"model":"zhipuai/glm-5.2","lab":"zhipuai","score":13,"best_input_per_m":0.3,"best_output_per_m":1.05,"points_per_dollar":14092.14}]},{"name":"LiveCodeBench","slug":"livecodebench","metric":"pass@1","entries":[{"model":"sakana/fugu-ultra","lab":"sakana","score":93.2,"best_input_per_m":5,"best_output_per_m":30,"points_per_dollar":4329.85},{"model":"sakana/fugu","lab":"sakana","score":92.9,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null},{"model":"nvidia/nemotron-3-ultra-550b-a55b","lab":"nvidia","score":89,"best_input_per_m":0.1,"best_output_per_m":0.1,"points_per_dollar":428915.66}]},{"name":"MLS-Bench-Lite","slug":"mls-bench-lite","metric":"score","entries":[{"model":"alibaba/qwen3.8-max-preview","lab":"alibaba","score":41,"best_input_per_m":2,"best_output_per_m":6,"points_per_dollar":7288.89},{"model":"moonshotai/kimi-k2.7-code","lab":"moonshotai","score":35.1,"best_input_per_m":0.275,"best_output_per_m":1.1,"points_per_dollar":35702.48}]},{"name":"AA-Briefcase","slug":"aa-briefcase","metric":"Elo","entries":[{"model":"anthropic/claude-opus-5","lab":"anthropic","score":1720,"best_input_per_m":5,"best_output_per_m":25,"points_per_dollar":89964.04},{"model":"moonshotai/kimi-k3","lab":"moonshotai","score":1548,"best_input_per_m":2,"best_output_per_m":8,"points_per_dollar":233660.38}]},{"name":"ARC-AGI-1","slug":"arc-agi-1","metric":"accuracy","entries":[{"model":"anthropic/claude-opus-5","lab":"anthropic","score":97.5,"best_input_per_m":5,"best_output_per_m":25,"points_per_dollar":5099.71},{"model":"openai/gpt-5.4-pro","lab":"openai","score":94.5,"best_input_per_m":27,"best_output_per_m":160,"points_per_dollar":771.27}]},{"name":"ARC-AGI-3","slug":"arc-agi-3","metric":"RHAE","entries":[{"model":"openai/gpt-6-astra","lab":"openai","score":99.9,"best_input_per_m":10,"best_output_per_m":50,"points_per_dollar":2625.49},{"model":"anthropic/claude-opus-5","lab":"anthropic","score":30.2,"best_input_per_m":5,"best_output_per_m":25,"points_per_dollar":1579.6}]},{"name":"MLE-Bench","slug":"mle-bench","metric":"average position score","entries":[{"model":"google/gemini-3.6-flash","lab":"google","score":63.9,"best_input_per_m":0.375,"best_output_per_m":1.875,"points_per_dollar":45632.67},{"model":"google/gemini-3.5-flash-lite","lab":"google","score":39.2,"best_input_per_m":0.15,"best_output_per_m":1.25,"points_per_dollar":48238.73}]},{"name":"JobBench","slug":"jobbench","metric":"success rate","entries":[{"model":"meta/muse-spark-1.1","lab":"meta","score":54.7,"best_input_per_m":1.25,"best_output_per_m":4.25,"points_per_dollar":14533.38},{"model":"moonshotai/kimi-k3","lab":"moonshotai","score":52.9,"best_input_per_m":2,"best_output_per_m":8,"points_per_dollar":7984.91}]},{"name":"MMLU-Pro","slug":"mmlu-pro","metric":"accuracy","entries":[{"model":"sakana/sakana-namazu","lab":"sakana","score":90.33,"best_input_per_m":0.95,"best_output_per_m":4,"points_per_dollar":27740.5},{"model":"nvidia/nemotron-3-ultra-550b-a55b","lab":"nvidia","score":86.8,"best_input_per_m":0.1,"best_output_per_m":0.1,"points_per_dollar":418313.25}]},{"name":"OmniDocBench","slug":"omnidocbench","metric":"overall edit distance","entries":[{"model":"openai/gpt-5.4-nano","lab":"openai","score":0.2419,"best_input_per_m":0.18,"best_output_per_m":1.1,"points_per_dollar":308.19},{"model":"openai/gpt-5.4-mini","lab":"openai","score":0.1263,"best_input_per_m":0.375,"best_output_per_m":4,"points_per_dollar":50.74}]},{"name":"OpenAI MRCR","slug":"openai-mrcr","metric":"accuracy","entries":[{"model":"openai/gpt-5.4-mini","lab":"openai","score":47.7,"best_input_per_m":0.375,"best_output_per_m":4,"points_per_dollar":19161.44},{"model":"openai/gpt-5.4-nano","lab":"openai","score":44.2,"best_input_per_m":0.18,"best_output_per_m":1.1,"points_per_dollar":56312.91}]},{"name":"Graphwalks","slug":"graphwalks","metric":"accuracy","entries":[{"model":"openai/gpt-5.4-mini","lab":"openai","score":76.3,"best_input_per_m":0.375,"best_output_per_m":4,"points_per_dollar":30650.26},{"model":"openai/gpt-5.4-nano","lab":"openai","score":73.4,"best_input_per_m":0.18,"best_output_per_m":1.1,"points_per_dollar":93515.1}]},{"name":"GeneBench","slug":"genebench","metric":"accuracy","entries":[{"model":"openai/gpt-5.5-pro","lab":"openai","score":33.2,"best_input_per_m":27.2727,"best_output_per_m":163.6364,"points_per_dollar":266.08},{"model":"openai/gpt-5.4-pro","lab":"openai","score":25.6,"best_input_per_m":27,"best_output_per_m":160,"points_per_dollar":208.94}]},{"name":"Terminal Bench 2.1","slug":"terminal-bench-2-1","metric":null,"entries":[{"model":"sakana/fugu-ultra","lab":"sakana","score":82.1,"best_input_per_m":5,"best_output_per_m":30,"points_per_dollar":453.09},{"model":"sakana/fugu","lab":"sakana","score":80.2,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null}]},{"name":"LiveCodeBench Pro","slug":"livecodebench-pro","metric":null,"entries":[{"model":"sakana/fugu-ultra","lab":"sakana","score":90.8,"best_input_per_m":5,"best_output_per_m":30,"points_per_dollar":4218.35},{"model":"sakana/fugu","lab":"sakana","score":87.8,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null}]},{"name":"τ3 Banking","slug":"3-banking","metric":null,"entries":[{"model":"sakana/fugu","lab":"sakana","score":21.7,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null},{"model":"sakana/fugu-ultra","lab":"sakana","score":20.6,"best_input_per_m":5,"best_output_per_m":30,"points_per_dollar":957.03}]},{"name":"Long Context Reasoning","slug":"long-context-reasoning","metric":null,"entries":[{"model":"sakana/fugu","lab":"sakana","score":74.7,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null},{"model":"sakana/fugu-ultra","lab":"sakana","score":73.3,"best_input_per_m":5,"best_output_per_m":30,"points_per_dollar":3405.34}]},{"name":"MRCRv2","slug":"mrcrv2","metric":null,"entries":[{"model":"sakana/fugu-ultra","lab":"sakana","score":93.6,"best_input_per_m":5,"best_output_per_m":30,"points_per_dollar":4348.43},{"model":"sakana/fugu","lab":"sakana","score":86.6,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null}]},{"name":"CTI-REALM","slug":"cti-realm","metric":null,"entries":[{"model":"sakana/fugu-ultra","lab":"sakana","score":69.4,"best_input_per_m":5,"best_output_per_m":30,"points_per_dollar":3224.16},{"model":"sakana/fugu","lab":"sakana","score":67.5,"best_input_per_m":null,"best_output_per_m":null,"points_per_dollar":null}]},{"name":"WideSearch","slug":"widesearch","metric":"F1","entries":[{"model":"alibaba/qwen3.8-max-preview","lab":"alibaba","score":81.9,"best_input_per_m":2,"best_output_per_m":6,"points_per_dollar":14560}]},{"name":"SWE-Bench Multimodal","slug":"swe-bench-multimodal","metric":"resolve rate","entries":[{"model":"anthropic/claude-opus-5","lab":"anthropic","score":59.4,"best_input_per_m":5,"best_output_per_m":25,"points_per_dollar":3106.9}]},{"name":"Frontier-Bench","slug":"frontier-bench","metric":"mean reward","entries":[{"model":"anthropic/claude-opus-5","lab":"anthropic","score":43.3,"best_input_per_m":5,"best_output_per_m":25,"points_per_dollar":2264.79}]},{"name":"DeepSearchQA","slug":"deepsearchqa","metric":"F1","entries":[{"model":"anthropic/claude-opus-5","lab":"anthropic","score":95,"best_input_per_m":5,"best_output_per_m":25,"points_per_dollar":4968.94}]},{"name":"HealthBench Professional","slug":"healthbench-professional","metric":"score","entries":[{"model":"anthropic/claude-opus-5","lab":"anthropic","score":59.8,"best_input_per_m":5,"best_output_per_m":25,"points_per_dollar":3127.82}]},{"name":"CyberGym","slug":"cybergym","metric":"score","entries":[{"model":"deepseek/deepseek-v4-flash-0731","lab":"deepseek","score":76.7,"best_input_per_m":0.035,"best_output_per_m":0.07,"points_per_dollar":938513.31}]},{"name":"DSBench-FullStack","slug":"dsbench-fullstack","metric":"score","entries":[{"model":"deepseek/deepseek-v4-flash-0731","lab":"deepseek","score":68.7,"best_input_per_m":0.035,"best_output_per_m":0.07,"points_per_dollar":840624.04}]},{"name":"DSBench-Hard","slug":"dsbench-hard","metric":"score","entries":[{"model":"deepseek/deepseek-v4-flash-0731","lab":"deepseek","score":59.6,"best_input_per_m":0.035,"best_output_per_m":0.07,"points_per_dollar":729275.01}]},{"name":"GDP.pdf","slug":"gdp-pdf","metric":"accuracy","entries":[{"model":"google/gemini-3.7-flash","lab":"google","score":34,"best_input_per_m":0.75,"best_output_per_m":3.75,"points_per_dollar":12141.34}]},{"name":"LMArena Text-to-Video Arena","slug":"lmarena-text-to-video-arena","metric":"Elo","entries":[{"model":"google/gemini-omni-flash-preview","lab":"google","score":1527,"best_input_per_m":1.5,"best_output_per_m":9,"points_per_dollar":222513.66}]},{"name":"IFEval","slug":"ifeval","metric":"accuracy","entries":[{"model":"meituan/longcat-2.0","lab":"meituan","score":90,"best_input_per_m":0.3,"best_output_per_m":1.2,"points_per_dollar":91435.54}]},{"name":"FORTE","slug":"forte","metric":"success rate","entries":[{"model":"meituan/longcat-2.0","lab":"meituan","score":73.2,"best_input_per_m":0.3,"best_output_per_m":1.2,"points_per_dollar":74367.57}]},{"name":"DeepSearch QA","slug":"deepsearch-qa","metric":null,"entries":[{"model":"meta/muse-glimmer-30b","lab":"meta","score":74.6,"best_input_per_m":0.2,"best_output_per_m":0.8,"points_per_dollar":111343.28}]},{"name":"AIME 2026","slug":"aime-2026","metric":"accuracy","entries":[{"model":"meta/muse-glimmer-30b","lab":"meta","score":94.7,"best_input_per_m":0.2,"best_output_per_m":0.8,"points_per_dollar":141343.28}]},{"name":"Finance Agent","slug":"finance-agent","metric":"accuracy","entries":[{"model":"meta/muse-spark-1.1","lab":"meta","score":57.2,"best_input_per_m":1.25,"best_output_per_m":4.25,"points_per_dollar":15197.61}]},{"name":"BabyVision","slug":"babyvision","metric":"accuracy","entries":[{"model":"meta/muse-spark-1.1","lab":"meta","score":76.3,"best_input_per_m":1.25,"best_output_per_m":4.25,"points_per_dollar":20272.33}]},{"name":"MMLU","slug":"mmlu","metric":"accuracy","entries":[{"model":"microsoft/phi-4-mini","lab":"microsoft","score":67.3,"best_input_per_m":0.075,"best_output_per_m":0.3,"points_per_dollar":251002.33}]},{"name":"τ³-Telecom","slug":"telecom","metric":"accuracy","entries":[{"model":"mistral/mistral-medium-2604","lab":"mistral","score":91.4,"best_input_per_m":1.5,"best_output_per_m":7.5,"points_per_dollar":16014.02}]},{"name":"Kimi Code Bench","slug":"kimi-code-bench","metric":null,"entries":[{"model":"moonshotai/kimi-k2.7-code","lab":"moonshotai","score":62,"best_input_per_m":0.275,"best_output_per_m":1.1,"points_per_dollar":63064.21}]},{"name":"MCP Mark Verified","slug":"mcp-mark-verified","metric":"success rate","entries":[{"model":"moonshotai/kimi-k2.7-code","lab":"moonshotai","score":81.1,"best_input_per_m":0.275,"best_output_per_m":1.1,"points_per_dollar":82492.05}]},{"name":"Kimi Claw 24/7 Bench","slug":"kimi-claw-24-7-bench","metric":null,"entries":[{"model":"moonshotai/kimi-k2.7-code","lab":"moonshotai","score":46.9,"best_input_per_m":0.275,"best_output_per_m":1.1,"points_per_dollar":47705.02}]},{"name":"SpreadsheetBench","slug":"spreadsheetbench","metric":"score","entries":[{"model":"moonshotai/kimi-k3","lab":"moonshotai","score":34.8,"best_input_per_m":2,"best_output_per_m":8,"points_per_dollar":5252.83}]},{"name":"ZeroBench","slug":"zerobench","metric":"pass@5","entries":[{"model":"moonshotai/kimi-k3","lab":"moonshotai","score":41,"best_input_per_m":2,"best_output_per_m":8,"points_per_dollar":6188.68}]},{"name":"GPQA","slug":"gpqa","metric":"accuracy","entries":[{"model":"nvidia/nemotron-3-ultra-550b-a55b","lab":"nvidia","score":87,"best_input_per_m":0.1,"best_output_per_m":0.1,"points_per_dollar":419277.11}]},{"name":"FinanceAgent","slug":"financeagent","metric":"accuracy","entries":[{"model":"openai/gpt-5.4-pro","lab":"openai","score":61.5,"best_input_per_m":27,"best_output_per_m":160,"points_per_dollar":501.94}]},{"name":"ScreenSpot-Pro","slug":"screenspot-pro","metric":"accuracy","entries":[{"model":"openai/gpt-6-astra","lab":"openai","score":92.7,"best_input_per_m":10,"best_output_per_m":50,"points_per_dollar":2436.27}]},{"name":"BenchCAD","slug":"benchcad","metric":"geometric overlap","entries":[{"model":"openai/gpt-6-astra","lab":"openai","score":95.9,"best_input_per_m":10,"best_output_per_m":50,"points_per_dollar":2520.37}]},{"name":"Terminal-Bench Science","slug":"terminal-bench-science","metric":"success rate","entries":[{"model":"openai/gpt-6-astra","lab":"openai","score":64.6,"best_input_per_m":10,"best_output_per_m":50,"points_per_dollar":1697.77}]},{"name":"ExploitBench","slug":"exploitbench","metric":"success rate","entries":[{"model":"openai/gpt-6-astra","lab":"openai","score":100,"best_input_per_m":10,"best_output_per_m":50,"points_per_dollar":2628.12}]},{"name":"SRE-Bench","slug":"sre-bench","metric":"success rate","entries":[{"model":"openai/gpt-6-astra","lab":"openai","score":88,"best_input_per_m":10,"best_output_per_m":50,"points_per_dollar":2312.75}]},{"name":"AIME26","slug":"aime26","metric":null,"entries":[{"model":"sakana/sakana-namazu","lab":"sakana","score":96.67,"best_input_per_m":0.95,"best_output_per_m":4,"points_per_dollar":29687.52}]},{"name":"LiveCodeBench v6","slug":"livecodebench-v6","metric":null,"entries":[{"model":"sakana/sakana-namazu","lab":"sakana","score":90.33,"best_input_per_m":0.95,"best_output_per_m":4,"points_per_dollar":27740.5}]},{"name":"JFBench","slug":"jfbench","metric":null,"entries":[{"model":"sakana/sakana-namazu","lab":"sakana","score":37.4,"best_input_per_m":0.95,"best_output_per_m":4,"points_per_dollar":11485.6}]},{"name":"Translation","slug":"translation","metric":null,"entries":[{"model":"sakana/sakana-namazu","lab":"sakana","score":52.2,"best_input_per_m":0.95,"best_output_per_m":4,"points_per_dollar":16030.71}]},{"name":"FairPoliticsQA","slug":"fairpoliticsqa","metric":null,"entries":[{"model":"sakana/sakana-namazu","lab":"sakana","score":56.3,"best_input_per_m":0.95,"best_output_per_m":4,"points_per_dollar":17289.83}]},{"name":"ClawEval","slug":"claweval","metric":"pass^3","entries":[{"model":"stepfun/step-3.7-flash","lab":"stepfun","score":67.1,"best_input_per_m":0.185,"best_output_per_m":1.11,"points_per_dollar":83668.44}]},{"name":"Terminal Bench 2.0","slug":"terminal-bench-2-0","metric":"score","entries":[{"model":"zhipuai/glm-4.7","lab":"zhipuai","score":33.4,"best_input_per_m":0.2,"best_output_per_m":0.8,"points_per_dollar":48759.12}]},{"name":"CritPt","slug":"critpt","metric":"accuracy","entries":[{"model":"zhipuai/glm-5.2","lab":"zhipuai","score":20.9,"best_input_per_m":0.3,"best_output_per_m":1.05,"points_per_dollar":22655.83}]},{"name":"AIME","slug":"aime","metric":"accuracy","entries":[{"model":"zhipuai/glm-5.2","lab":"zhipuai","score":99.2,"best_input_per_m":0.3,"best_output_per_m":1.05,"points_per_dollar":107533.88}]},{"name":"HMMT","slug":"hmmt","metric":"accuracy","entries":[{"model":"zhipuai/glm-5.2","lab":"zhipuai","score":94.4,"best_input_per_m":0.3,"best_output_per_m":1.05,"points_per_dollar":102330.62}]},{"name":"IMOAnswerBench","slug":"imoanswerbench","metric":"accuracy","entries":[{"model":"zhipuai/glm-5.2","lab":"zhipuai","score":91,"best_input_per_m":0.3,"best_output_per_m":1.05,"points_per_dollar":98644.99}]},{"name":"PostTrainBench","slug":"posttrainbench","metric":"score","entries":[{"model":"zhipuai/glm-5.2","lab":"zhipuai","score":34.3,"best_input_per_m":0.3,"best_output_per_m":1.05,"points_per_dollar":37181.57}]},{"name":"Tool-Decathlon","slug":"tool-decathlon","metric":"score","entries":[{"model":"zhipuai/glm-5.2","lab":"zhipuai","score":48.2,"best_input_per_m":0.3,"best_output_per_m":1.05,"points_per_dollar":52249.32}]}]}