Model
Claude Opus 5
Strongest Claude Opus model for coding, agents, and professional work
Compare providers (sorted by listed input price)
30 of 33 providers publish a price for Claude Opus 5. The cheapest full-context option is NanoGPT at $5/$25 per M. Listed prices span 2.0× for identical weights.
Best price history
Daily lowest published price across live providers. A lower line is better for the buyer.
- Best input /M
- $5▲ 11.1%
- Best output /M
- $25▲ 11.1%
- Provider coverage
- 43
- 39 at the start · 24 snapshots
Price badge
https://www.vaanalytics.in/badge/anthropic/claude-opus-5.svgLive SVG, regenerated on every hourly sync — paste it into a README to always show the current cheapest listed price. Updates when prices change; no build step needed on your side.
Recent changes
Changelog →- new modelvia pioneer · Sep 13, 2026
- new modelvia 302ai · Sep 9, 2026
- contextvia cortecs · Sep 2, 2026
- output: 1M → 128K tokens
- new modelvia kenari · Aug 28, 2026
- repricedvia requesty · Aug 28, 2026
- input: $4.50 → $5
- output: $22.50 → $25
- cache_read: $0.45 → $0.5
- cache_write: $5.63 → $6.25
- repricedvia requesty · Aug 28, 2026
- input: $4.95 → $5.50
- output: $24.75 → $27.50
- cache_read: $0.495 → $0.55
- cache_write: $6.19 → $6.88
Benchmarks
- SWE-Bench Verified ↗(resolved)report ↗Unclassified source96
- SWE-Bench Pro ↗(resolve rate)report ↗Unclassified source79.2
- SWE-Bench Multilingual ↗(resolve rate)report ↗Unclassified source89.5
- SWE-Bench Multimodal ↗(resolve rate)report ↗Unclassified source59.4
- DeepSWE(resolve rate)report ↗Unclassified source68.8
- FrontierCode(mean@5)report ↗Unclassified source53.4
- Frontier-Bench(mean reward)report ↗Unclassified source43.3
- BrowseComp ↗(accuracy)report ↗Unclassified source90.8
- Humanity's Last Exam ↗(accuracy)report ↗Unclassified source56.3
- Humanity's Last Exam ↗(accuracy)report ↗Unclassified source64.7
- DeepSearchQA(F1)report ↗Unclassified source95
- OSWorld ↗(success rate)report ↗Unclassified source70.6
- GDPval-AA(Elo)report ↗Unclassified source1861
- AA-Briefcase(Elo)report ↗Unclassified source1720
- AutomationBench(success rate)report ↗Unclassified source26
- ARC-AGI-1 ↗(accuracy)report ↗Unclassified source97.5
- ARC-AGI-2 ↗(accuracy)report ↗Unclassified source90.4
- ARC-AGI-3 ↗(RHAE)report ↗Unclassified source30.2
- HealthBench Professional(score)report ↗Unclassified source59.8
Scores a lab published about its own model are claims, not measurements. Rankings elsewhere on this site use independently measured scores only.
Prices are per million tokens (USD). “—” means the provider does not publicly list a price for this model. Data from models.dev; verify with the provider before purchasing.