OpenAIGPT-5.1-Codex
GPT-5.1-Codex is an OpenAI proprietary reasoning model, released 12 Nov 2025. Its best configuration (high reasoning effort) ranks #209 of 785 on the BenchLeader Index at 56.3 ±4.2, in the upper half. The ± is the point: 270 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in instruction following (69) and lowest in reasoning (50). At $3.44 per million tokens blended it is among the most expensive ranked models. Output speed of 127 tokens per second puts it faster than most, with a first answer in 8.1 s. It has been measured at 3 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 13 Nov 2025.stale: no new result in six months
- Blended price
- $3.44/M
- $1.25 in · $10.00 out
- Output speed
- 127 tok/s
- measured by Artificial Analysis
- First answer
- 8.06 s
- first token 4.87 s
- Context
- 400k
- Full answer
- 12 s
- median, reasoning included
- Cost per run
- –
- Released
- 12 Nov 2025
- Overall index56
- Reasoning50
- Agents & tools64
- Knowledge59
- Instruction following69
- Multimodal55
- Long context60
- Composite56
Versions
OpenAI has shipped 5 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| GPT-5.3-Codexxhigh | 5 Feb 2026 | 63.8 | #60 |
| GPT-5.2-Codexxhigh | 18 Dec 2025 | 59.4 | #133 |
| GPT-5.1-Codexhighthis page | 12 Nov 2025 | 56.3 | #209 |
| GPT-5-Codexhigh | 15 Sept 2025 | 57.5 | #177 |
| GPT 5.5 Codex | – | – | – |
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| medium | – | – | 65 tok/s | 4.87 s | $0.0035 | Coding 59 |
| highbest | 56.3 | #209 | 127 tok/s | 8.06 s | $0.0035 | Agents & tools 64 · Composite 56 · Instruction following 69 · Knowledge 59 · Long context 60 · Multimodal 55 · Reasoning 50 |
| not stated | – | – | 65 tok/s | 4.87 s | $0.0035 | Agents & tools 61 · Coding 50 |
How its index has moved
19 Sept 2026 to 21 Oct 2026The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | high | Source | Trend |
|---|---|---|---|
| GPQA Diamond (AA)not in index | 86.0%#116 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 25.7%#182 | Artificial Analysis | |
| CritPt | 5.7%#148 | Artificial Analysis |
Coding
| Benchmark | medium | high | not stated | Source | Trend |
|---|---|---|---|---|---|
| Arena WebDeveffort not stated | – | – | 1337#114 | Arena (LMArena) | |
| LiveCodeBencheffort not stated | – | – | 85.5%#30 | Vals AI | |
| SWE-bench Verified (bash only) | 66.0%#17 | – | – | SWE-bench | |
| SWE-bench Verified (any scaffold)not in index | 66.0%#19 | – | – | SWE-bench | |
| Vibe Code Bench v1.1not in index | – | 13.1%#94 | – | Vals AI | |
| ALE-Benchnot in indexeffort not stated | – | – | 1244.9#38 | Epoch AI Benchmarking Hub |
Agents & tools
| Benchmark | high | not stated | Source | Trend |
|---|---|---|---|---|
| Terminal-Bencheffort not stated | – | 57.8%#23 | Terminal-Bench | |
| Terminal-Bench Hard | 34.9%#82 | – | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 83.0%#113 | – | Artificial Analysis |
Knowledge
| Benchmark | high | Source | Trend |
|---|---|---|---|
| AA-Omniscience | -6.5#168 | Artificial Analysis | |
| AA-Omniscience: accuracynot in index | 39.9%#125 | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 22.8%#261 | Artificial Analysis | |
| Global-MMLU-Lite (AA)not in index | 89.7%#12 | Artificial Analysis |
Instruction following
| Benchmark | high | Source | Trend |
|---|---|---|---|
| IFBench | 70.0%#70 | Artificial Analysis |
Multimodal
| Benchmark | high | Source | Trend |
|---|---|---|---|
| MMMU-Pro | 72.5%#133 | Artificial Analysis |
Long context
| Benchmark | high | Source | Trend |
|---|---|---|---|
| AA-LCR | 69.3%#226 | Artificial Analysis |
Composite
| Benchmark | high | Source | Trend |
|---|---|---|---|
| AA Intelligence Index v4.3.2 | 23.7#183 | Artificial Analysis |
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| Azure | 49 tok/s | 1.53 s | $1.25 | $10.00 | 400k | – |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.125 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0035 | $0.0032 | 10.4 s |
| Summarize a 30-page report | 12,000 / 600 | $0.021 | $0.011 | 12.8 s |
| Code edit | 6,000 / 1,500 | $0.022 | $0.017 | 19.9 s |
| Agentic coding session | 60,000 / 4,000 | $0.115 | $0.064 | 39.5 s |
| Structured extraction | 2,000 / 200 | $0.0045 | $0.0028 | 9.6 s |
More from OpenAI
See also
Questions people ask
- What are GPT-5.1-Codex's benchmark scores?
- As of 25 Oct 2026, GPT-5.1-Codex (high effort) scores 56.3 ±4.2 on the BenchLeader Index, strongest in instruction following (69) and weakest in reasoning (50). That rests on 7 independent benchmark results, and every score on this page links to its publisher.
- How does GPT-5.1-Codex rank?
- GPT-5.1-Codex (high effort) is #209 of 785 ranked configurations on 25 Oct 2026. 270 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
- How much does GPT-5.1-Codex cost per 1M tokens?
- As of 25 Oct 2026, GPT-5.1-Codex costs $1.25 per 1M input tokens and $10.00 per 1M output tokens, $3.44 blended (three input tokens per output token).
- When was GPT-5.1-Codex released?
- OpenAI released GPT-5.1-Codex on 12 Nov 2025. BenchLeader has results for it at 3 reasoning-effort settings.
Data as of 25 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “GPT-5.1-Codex: benchmarks, pricing, speed and rank”, data as of 25 Oct 2026, https://www.benchleader.com/models/gpt-5-1-codex.