OpenAIo1
o1 is an OpenAI proprietary reasoning model, released 17 Dec 2024. Its best configuration ranks #299 of 785 on the BenchLeader Index at 53.1 ±4.1, in the upper half. The ± is the point: 290 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 agents & tools (42). At $26.25 per million tokens blended it is among the most expensive ranked models. It has been measured at 4 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 22 Oct 2026.
- Blended price
- $26.25/M
- $15.00 in · $60.00 out
- Output speed
- –
- First answer
- –
- Context
- 200k
- Full answer
- –
- Cost per run
- –
- Released
- 17 Dec 2024
- Overall index53
- Reasoning42
- Coding56
- Agents & tools42
- Maths54
- Knowledge57
- Instruction following69
- Human preference58
- Multimodal54
- Long context57
- Composite47
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 |
|---|---|---|---|---|---|---|
| low | – | – | – | – | $0.024 | Maths 36 · Reasoning 54 |
| medium | 48.1 | #436 | – | – | $0.024 | Long context 48 · Maths 46 · Reasoning 48 |
| high | 48.3 | #430 | – | – | $0.024 | Coding 39 · Knowledge 54 · Maths 49 · Reasoning 49 |
| not statedbest | 53.1 | #299 | – | – | $0.024 | Agents & tools 42 · Coding 56 · Composite 47 · Human preference 58 · Instruction following 69 · Knowledge 57 · Long context 57 · Maths 54 · Multimodal 54 · Reasoning 42 |
How its index has moved
8 Sept 2026 to 21 Oct 2026- not stated
- high
- medium
The 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 | low | medium | high | not stated | Source | Trend |
|---|---|---|---|---|---|---|
| GPQA Diamond | 74.2%#154 | 75.8%#144 | 76.8%#140 | 50.3%#215 | Epoch AI Benchmarking Hub | |
| Humanity's Last Exameffort not stated | – | – | – | 8.0%#36 | Scale AI / CAIS | |
| SimpleBench | – | 36.7%#62 | 40.1%#60 | 41.7%#56 | SimpleBench | |
| Arena Hard Promptseffort not stated | – | – | – | 1418#153 | Arena (LMArena) | |
| GPQA Diamond (AA)not in indexeffort not stated | – | – | – | 76.5%#234 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in indexeffort not stated | – | – | – | 7.0%#374 | Artificial Analysis | |
| GPQA Diamond (Vals)not in index | – | – | 73.2%#82 | – | Vals AI | |
| CritPteffort not stated | – | – | – | 0.3%#271 | Artificial Analysis | |
| Chess Puzzlesnot in index | 9.0%#122 | 8.0%#124 | 10.0%#119 | – | Epoch AI Benchmarking Hub | |
| LMCAnot in index | – | – | 22.3%#155 | – | Epoch AI Benchmarking Hub | |
| DTBenchnot in index | – | – | 74.7%#151 | – | Epoch AI Benchmarking Hub |
Coding
| Benchmark | medium | high | not stated | Source | Trend |
|---|---|---|---|---|---|
| WeirdML | – | 46.1%#81 | 47.6%#76 | WeirdML | |
| Arena Codingeffort not stated | – | – | 1433#168 | Arena (LMArena) | |
| LiveCodeBench | – | 50.3%#111 | – | Vals AI | |
| Aider Polygloteffort not stated | – | – | 61.7%#12 | Aider polyglot leaderboard | |
| SWE-bench Verified (any scaffold)not in indexeffort not stated | – | – | 64.6%#22 | SWE-bench | |
| CadEvalnot in index | 56.0%#4 | – | – | Epoch AI Benchmarking Hub |
Agents & tools
| Benchmark | medium | not stated | Source | Trend |
|---|---|---|---|---|
| Cybencheffort not stated | – | 10.0%#15 | Cybench | |
| Terminal-Bench Hardeffort not stated | – | 12.9%#206 | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in indexeffort not stated | – | 62.6%#175 | Artificial Analysis | |
| METR Time Horizonsnot in index | 51.0%#26 | 45.1%#29 | METR | |
| BrowseCompnot in indexeffort not stated | – | 6.3%#5 | OpenAI |
Maths
| Benchmark | low | medium | high | not stated | Source | Trend |
|---|---|---|---|---|---|---|
| FrontierMath Tiers 1–3 | 8.4%#107 | 10.2%#105 | 14.7%#102 | – | Epoch AI Benchmarking Hub | |
| OTIS Mock AIME | 53.3%#178 | 73.3%#129 | 73.3%#129 | 31.1%#205 | Epoch AI Benchmarking Hub | |
| MATH Level 5 | – | 94.4%#17 | 94.7%#16 | 81.7%#33 | Epoch AI Benchmarking Hub | |
| AIME (Vals) | – | – | 71.5%#53 | – | Vals AI | |
| MGSM | – | – | 89.3%#44 | – | Vals AI | |
| MATH 500not in index | – | – | 90.4%#24 | – | Vals AI | |
| Arena Mathseffort not stated | – | – | – | 1409#136 | Arena (LMArena) |
Knowledge
| Benchmark | high | not stated | Source | Trend |
|---|---|---|---|---|
| SimpleQA Verified | 41.1%#45 | – | Epoch AI Benchmarking Hub | |
| AA-Omniscienceeffort not stated | – | -11.1#206 | Artificial Analysis | |
| MMLU-Pro | 83.5%#70 | – | Vals AI | |
| LegalBench | 80.4%#85 | – | Vals AI | |
| TaxEval | 74.3%#38 | – | Vals AI | |
| MedQA | 96.5%#1 | – | Vals AI | |
| AA-Omniscience: accuracynot in indexeffort not stated | – | 34.5%#159 | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in indexeffort not stated | – | 30.4%#221 | Artificial Analysis |
Instruction following
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| IFBench | 70.3%#68 | Artificial Analysis | |
| Arena Instruction Followingnot in index | 1407#128 | Arena (LMArena) |
Human preference
| Benchmark | medium | not stated | Source | Trend |
|---|---|---|---|---|
| Arena Texteffort not stated | – | 1402#146 | Arena (LMArena) | |
| Arena Creative Writingnot in indexeffort not stated | – | 1381#135 | Arena (LMArena) | |
| Arena Multi-turnnot in indexeffort not stated | – | 1390#166 | Arena (LMArena) | |
| Arena Longer Queriesnot in indexeffort not stated | – | 1412#143 | Arena (LMArena) | |
| Lech Mazur Creative Writingnot in index | 70.2%#33 | – | Lech Mazur |
Multimodal
| Benchmark | medium | high | not stated | Source | Trend |
|---|---|---|---|---|---|
| Arena Visioneffort not stated | – | – | 1168#108 | Arena (LMArena) | |
| VISTAeffort not stated | – | – | 45.3%#28 | Scale AI SEAL | |
| MMMU (validation)effort not stated | – | – | 78.2%#9 | MMMU | |
| MMMU-Pro (Vals)not in index | – | 77.4%#51 | – | Vals AI | |
| GeoBenchnot in index | 80.0%#5 | – | – | GeoBench | |
| VPCTnot in index | 37.0%#24 | – | – | Epoch AI Benchmarking Hub | |
| SpatialViz-Benchnot in indexeffort not stated | – | – | 41.4%#2 | Epoch AI Benchmarking Hub |
Long context
| Benchmark | medium | not stated | Source | Trend |
|---|---|---|---|---|
| Fiction.LiveBench 120k | 53.1%#20 | – | Fiction.live | |
| AA-LCReffort not stated | – | 65.0%#252 | Artificial Analysis |
Composite
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Epoch Capabilities Indexnot in index | 141.7#105 | Epoch AI Benchmarking Hub | |
| AA Intelligence Index v4.3.2 | 15.2#286 | Artificial Analysis |
Safety & honesty
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| FORTRESSnot in index | 19.4%#41 | Scale AI SEAL | |
| MASKnot in index | 59.3%#39 | Scale AI SEAL |
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 $7.50 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.024 | $0.022 | – |
| Summarize a 30-page report | 12,000 / 600 | $0.216 | $0.148 | – |
| Code edit | 6,000 / 1,500 | $0.180 | $0.146 | – |
| Agentic coding session | 60,000 / 4,000 | $1.14 | $0.802 | – |
| Structured extraction | 2,000 / 200 | $0.042 | $0.031 | – |
See also
Questions people ask
- What are o1's benchmark scores?
- As of 24 Oct 2026, o1 scores 53.1 ±4.1 on the BenchLeader Index, strongest in instruction following (69) and weakest in agents & tools (42). That rests on 21 independent benchmark results, and every score on this page links to its publisher.
- How does o1 rank?
- o1 is #299 of 785 ranked configurations on 24 Oct 2026. 290 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 o1 cost per 1M tokens?
- As of 24 Oct 2026, o1 costs $15.00 per 1M input tokens and $60.00 per 1M output tokens, $26.25 blended (three input tokens per output token).
- When was o1 released?
- OpenAI released o1 on 17 Dec 2024. BenchLeader has results for it at 4 reasoning-effort settings.
Data as of 24 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “o1: benchmarks, pricing, speed and rank”, data as of 24 Oct 2026, https://www.benchleader.com/models/o1.