DeepSeek-V4-Pro-0813 vs GPT-5.1-Codex-mini
DeepSeek-V4-Pro-0813 (DeepSeek) and GPT-5.1-Codex-mini (OpenAI) compared on benchmarks, pricing, context window, and use-case rankings.
- GPT-5.1-Codex-mini is cheaper on input tokens — $0.25 vs $0.58 per 1M.
- DeepSeek-V4-Pro-0813 has a larger context window (1M vs 400K tokens).
- DeepSeek-V4-Pro-0813 ranks higher for coding (77th vs 45th percentile).
| Attribute | DeepSeek-V4-Pro-0813 | GPT-5.1-Codex-mini |
|---|---|---|
| Scores | ||
| Intelligence (ECI) | 155 | — |
| Coding | 77 | 45 |
| Math | 88 | 57 |
| Reasoning & Knowledge | 69 | 23 |
| Agentic & Tools | — | 42 |
| Specifications | ||
| Developer | DeepSeek | OpenAI |
| Family | DeepSeek | GPT |
| Released | Aug 13, 2026 | Nov 12, 2025 |
| Parameters | 1.6T | — |
| Availability | Open weights (unrestricted) | API access |
| Context window | 1M | 400K |
| Price — $/M input | $0.58 | $0.25 |
| Price — $/M output | $1.74 | $2.00 |
| Inputs | text | image, text |
| Outputs | text | text |
| Benchmarks | ||
| AIME 2024/2025 | 99% | 92% |
| ARC-AGI | 91% | — |
| ARC-AGI-2 | 61% | — |
| GPQA Diamond | 92% | 81% |
| Humanity's Last Exam | 41% | 19% |
| SciCode | 51% | 43% |
| SimpleQA Verified | 53% | — |
| WebDev Arena | 1582 | 1247 |
| WeirdML | 66% | — |
| LiveCodeBench | — | 84% |
| MMLU-Pro | — | 82% |
| Terminal-Bench | — | 62% |
| τ²-bench | — | 63% |
Use-case scores are 0–100 percentile composites across each area’s benchmarks, ranked against every model from the past year. Highlighted cells lead each row. Open a model for the full picture.
Frequently asked questions
Which is cheaper, DeepSeek-V4-Pro-0813 or GPT-5.1-Codex-mini?
GPT-5.1-Codex-mini is cheaper on input tokens at $0.25 per million, versus $0.58 (representative OpenRouter pricing).
Which has a larger context window, DeepSeek-V4-Pro-0813 or GPT-5.1-Codex-mini?
DeepSeek-V4-Pro-0813 supports up to 1M tokens, compared with 400K for the other.
Which is better for coding, DeepSeek-V4-Pro-0813 or GPT-5.1-Codex-mini?
Across coding benchmarks like SWE-bench Verified and Terminal-Bench, DeepSeek-V4-Pro-0813 ranks higher — 77th vs 45th percentile among the models tracked on Model Beat.
Want a different match-up? Open the compare tool to add or swap models.
More comparisons
Benchmarks & model data from Epoch AI (CC BY); pricing & specs from OpenRouter. ECI = Epoch Capabilities Index.