Model
DeepSeek V4 Pro 0813
DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes
Compare providers (sorted by listed input price)
27 of 32 providers publish a price for DeepSeek V4 Pro 0813. The cheapest full-context option is RunInfra at $0.6/$1.9 per M. OrcaRouter lists lower but caps context at 1M.
Best price history
Daily lowest published price across live providers. A lower line is better for the buyer.
- Best input /M
- $0.442▼ 26.6%
- Best output /M
- $0.884▼ 51.1%
- Provider coverage
- 40
- 26 at the start · 24 snapshots
Price badge
https://www.vaanalytics.in/badge/deepseek/deepseek-v4-pro-0813.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 →- repricedvia edenai · Sep 14, 2026
- input: $1.12 → $0.581
- output: $3.37 → $1.74
- contextvia kilo · Sep 14, 2026
- context: 1.05M → 1.02M tokens
- output: 393K → 384K tokens
- repricedvia openrouter · Sep 14, 2026
- output: 393K → 384K tokens
- input: $0.578 → $1.05
- output: $1.73 → $3.16
- cache_read: $0.018 → $0.035
- repricedvia openrouter · Sep 13, 2026
- output: 384K → 393K tokens
- input: $1.05 → $0.578
- output: $3.15 → $1.73
- cache_read: $0.035 → $0.018
- contextvia kilo · Sep 13, 2026
- context: 1.02M → 1.05M tokens
- output: 384K → 393K tokens
- repricedvia edenai · Sep 13, 2026
- input: $0.581 → $1.12
- output: $1.74 → $3.37
Weights
Standard open-source terms. Commercial use without additional conditions. Weights download without an access request.
licence and gating from the Hugging Face model card ↗ — the card is authoritative, this is a summary
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.