OpenAIGPT-5-Codex
GPT-5-Codex is an OpenAI proprietary reasoning model, released 15 Sept 2025. Its best configuration (high reasoning effort) ranks #177 of 785 on the BenchLeader Index at 57.5 ±4.3, in the upper half. The ± is the point: 258 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 (72) and lowest in reasoning (49). At $3.44 per million tokens blended it is among the most expensive ranked models. Output speed of 172 tokens per second puts it in the fastest quarter, with a first answer in 10.3 s. It has been measured at 2 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 1 Sept 2026.
- Blended price
- $3.44/M
- $1.25 in · $10.00 out
- Output speed
- 172 tok/s
- measured by Artificial Analysis
- First answer
- 10 s
- Context
- 400k
- Full answer
- 13 s
- median, reasoning included
- Cost per run
- –
- Released
- 15 Sept 2025
- Overall index58
- Reasoning49
- Coding59
- Agents & tools67
- Knowledge58
- Instruction following72
- Multimodal57
- Long context61
- Composite57
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-Codexhigh | 12 Nov 2025 | 56.3 | #209 |
| GPT-5-Codexhighthis page | 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 |
|---|---|---|---|---|---|---|
| highbest | 57.5 | #177 | 172 tok/s | 10 s | $0.0035 | Agents & tools 67 · Coding 59 · Composite 57 · Instruction following 72 · Knowledge 58 · Long context 61 · Multimodal 57 · Reasoning 49 |
| not stated | – | – | – | – | $0.0035 | Agents & tools 51 · Coding 53 · Reasoning 64 |
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 | not stated | Source | Trend |
|---|---|---|---|---|
| GPQA Diamond (AA)not in index | 83.7%#155 | – | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 27.9%#171 | – | Artificial Analysis | |
| Kagi LLM Benchmarkeffort not stated | – | 70.3%#26 | Kagi LLM Benchmark | |
| CritPt | 5.1%#154 | – | Artificial Analysis |
Coding
| Benchmark | high | not stated | Source | Trend |
|---|---|---|---|---|
| WeirdML | 54.5%#59 | 54.5%#59 | WeirdML | |
| LiveCodeBench | 84.7%#35 | – | Vals AI |
Agents & tools
| Benchmark | high | not stated | Source | Trend |
|---|---|---|---|---|
| Terminal-Bencheffort not stated | – | 44.3%#43 | Terminal-Bench | |
| Terminal-Bench Hard | 37.9%#62 | – | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 86.8%#83 | – | Artificial Analysis |
Knowledge
| Benchmark | high | Source | Trend |
|---|---|---|---|
| AA-Omniscience | -7.5#175 | Artificial Analysis | |
| AA-Omniscience: accuracynot in index | 38.7%#132 | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 24.6%#247 | Artificial Analysis |
Instruction following
| Benchmark | high | Source | Trend |
|---|---|---|---|
| IFBench | 74.2%#36 | Artificial Analysis |
Multimodal
| Benchmark | high | Source | Trend |
|---|---|---|---|
| MMMU-Pro | 73.8%#120 | Artificial Analysis |
Long context
| Benchmark | high | Source | Trend |
|---|---|---|---|
| AA-LCR | 71.7%#196 | Artificial Analysis |
Composite
| Benchmark | high | Source | Trend |
|---|---|---|---|
| AA Intelligence Index v4.3.2 | 24.9#167 | Artificial Analysis |
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 | 12.0 s |
| Summarize a 30-page report | 12,000 / 600 | $0.021 | $0.011 | 13.7 s |
| Code edit | 6,000 / 1,500 | $0.022 | $0.017 | 19.0 s |
| Agentic coding session | 60,000 / 4,000 | $0.115 | $0.064 | 33.5 s |
| Structured extraction | 2,000 / 200 | $0.0045 | $0.0028 | 11.4 s |
More from OpenAI
See also
Questions people ask
- What are GPT-5-Codex's benchmark scores?
- As of 25 Oct 2026, GPT-5-Codex (high effort) scores 57.5 ±4.3 on the BenchLeader Index, strongest in instruction following (72) and weakest in reasoning (49). That rests on 9 independent benchmark results, and every score on this page links to its publisher.
- How does GPT-5-Codex rank?
- GPT-5-Codex (high effort) is #177 of 785 ranked configurations on 25 Oct 2026. 258 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-Codex cost per 1M tokens?
- As of 25 Oct 2026, GPT-5-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-Codex released?
- OpenAI released GPT-5-Codex on 15 Sept 2025. BenchLeader has results for it at 2 reasoning-effort settings.
Data as of 25 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “GPT-5-Codex: benchmarks, pricing, speed and rank”, data as of 25 Oct 2026, https://www.benchleader.com/models/gpt-5-codex.