OpenAIo4-mini
o4-mini is an OpenAI proprietary reasoning model, released 16 Apr 2025. Its best configuration ranks #192 of 785 on the BenchLeader Index at 56.9 ±5.9, in the upper half. The ± is the point: 299 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 maths (69) and lowest in knowledge (38). At $1.93 per million tokens blended it is pricier than most 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
- $1.93/M
- $1.10 in · $4.40 out
- Output speed
- –
- First answer
- –
- Context
- 200k
- Full answer
- –
- Cost per run
- –
- Released
- 16 Apr 2025
- Overall index57
- Reasoning59
- Coding55
- Agents & tools64
- Maths69
- Knowledge38
- Instruction following69
- Human preference56
- Multimodal57
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 | 41.6 | #625 | – | – | $0.0018 | Knowledge 32 · Maths 39 · Reasoning 41 |
| medium | 51.9 | #333 | – | – | $0.0018 | Long context 54 · Maths 46 · Multimodal 66 · Reasoning 44 |
| high | 50.1 | #380 | 144 tok/s | 25 s | $0.0018 | Agents & tools 43 · Coding 52 · Composite 48 · Instruction following 50 · Knowledge 47 · Long context 55 · Maths 51 · Multimodal 59 · Reasoning 45 |
| not statedbest | 56.9 | #192 | – | – | $0.0018 | Agents & tools 64 · Coding 55 · Human preference 56 · Instruction following 69 · Knowledge 38 · Maths 69 · Multimodal 57 · Reasoning 59 |
How its index has moved
8 Sept 2026 to 23 Oct 2026- not stated
- medium
- high
- low
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
Coding
| Benchmark | medium | high | not stated | Source | Trend |
|---|---|---|---|---|---|
| WeirdML | – | 52.6%#64 | – | WeirdML | |
| GSO-Bench | – | 3.6%#31 | – | GSO-Bench | |
| Arena Codingeffort not stated | – | – | 1433#166 | Arena (LMArena) | |
| LiveCodeBench | – | 82.2%#53 | – | Vals AI | |
| Aider Polygloteffort not stated | – | – | 72.0%#7 | Aider polyglot leaderboard | |
| SWE-bench Verified (bash only)effort not stated | – | – | 45.0%#34 | SWE-bench | |
| SWE-bench Verified (any scaffold)not in indexeffort not stated | – | – | 64.6%#22 | SWE-bench | |
| CadEvalnot in index | 62.0%#3 | – | – | Epoch AI Benchmarking Hub | |
| ALE-Benchnot in index | – | 826.2#69 | – | Epoch AI Benchmarking Hub | |
| AlgoTunenot in index | – | 1.7#6 | – | AlgoTune |
Agents & tools
| Benchmark | medium | high | not stated | Source | Trend |
|---|---|---|---|---|---|
| GDPval | – | 25.3%#8 | – | OpenAI | |
| Terminal-Bench Hard | – | 15.2%#193 | – | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | – | 55.6%#185 | – | Artificial Analysis | |
| BFCL Overalleffort not stated | – | – | 53.2%#19 | Berkeley Function Calling Leaderboard | |
| METR Time Horizonsnot in index | 63.9%#16 | – | – | METR | |
| BrowseCompnot in indexeffort not stated | – | – | 3.3%#7 | OpenAI |
Maths
| Benchmark | low | medium | high | not stated | Source | Trend |
|---|---|---|---|---|---|---|
| FrontierMath Tiers 1–3 | 16.1%#101 | 28.8%#76 | 36.1%#66 | – | Epoch AI Benchmarking Hub | |
| FrontierMath Tier 4 | – | – | 4.9%#66 | – | Epoch AI Benchmarking Hub | |
| OTIS Mock AIME | 57.8%#167 | 73.3%#129 | 81.7%#112 | – | Epoch AI Benchmarking Hub | |
| MATH Level 5 | – | – | 97.8%#4 | – | Epoch AI Benchmarking Hub | |
| AIME (Vals) | – | – | 83.7%#41 | – | Vals AI | |
| MGSM | – | – | 93.4%#9 | – | Vals AI | |
| IMO 2025 | – | – | 14.3%#5 | – | MathArena | |
| Omni-MATH (HELM)effort not stated | – | – | – | 72.0%#2 | HELM Capabilities | |
| MATH 500not in index | – | – | 94.2%#10 | – | Vals AI | |
| Arena Mathseffort not stated | – | – | – | 1418#124 | Arena (LMArena) |
Knowledge
| Benchmark | low | high | not stated | Source | Trend |
|---|---|---|---|---|---|
| SimpleQA Verified | 19.6%#75 | 18.8%#77 | – | Epoch AI Benchmarking Hub | |
| AA-Omniscience | – | -35.7#303 | – | Artificial Analysis | |
| MMLU-Pro | – | 80.6%#82 | – | Vals AI | |
| LegalBench | – | 79.2%#94 | – | Vals AI | |
| CorpFin | – | 59.0%#77 | – | Vals AI | |
| TaxEval | – | 74.8%#28 | – | Vals AI | |
| MedQA | – | 96.0%#9 | – | Vals AI | |
| PRBench Financeeffort not stated | – | – | 39.2%#27 | Scale AI SEAL | |
| PRBench Legaleffort not stated | – | – | 38.1%#36 | Scale AI SEAL | |
| MultiNRC | – | 22.2%#32 | – | Scale AI SEAL | |
| MMLU-Pro (HELM)not in indexeffort not stated | – | – | 82.0%#13 | HELM Capabilities | |
| MedCodenot in index | – | 33.8%#81 | – | Vals AI | |
| MedScribenot in index | – | 69.1%#93 | – | Vals AI | |
| AA-Omniscience: accuracynot in index | – | 24.8%#238 | – | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | – | 19.6%#285 | – | Artificial Analysis |
Instruction following
| Benchmark | high | not stated | Source | Trend |
|---|---|---|---|---|
| IFBench | 68.7%#79 | – | Artificial Analysis | |
| MultiChallenge | 44.9%#29 | – | Scale AI SEAL | |
| IFEval (HELM)effort not stated | – | 92.9%#5 | HELM Capabilities | |
| WildBench (HELM)effort not stated | – | 85.4%#9 | HELM Capabilities | |
| Arena Instruction Followingnot in indexeffort not stated | – | 1368#174 | Arena (LMArena) |
Human preference
| Benchmark | medium | not stated | Source | Trend |
|---|---|---|---|---|
| Arena Texteffort not stated | – | 1391#160 | Arena (LMArena) | |
| Arena Creative Writingnot in indexeffort not stated | – | 1337#179 | Arena (LMArena) | |
| Arena Multi-turnnot in indexeffort not stated | – | 1383#172 | Arena (LMArena) | |
| Arena Longer Queriesnot in indexeffort not stated | – | 1370#188 | Arena (LMArena) | |
| Lech Mazur Creative Writingnot in index | 75.0%#26 | – | Lech Mazur |
Multimodal
| Benchmark | medium | high | not stated | Source | Trend |
|---|---|---|---|---|---|
| Arena Visioneffort not stated | – | – | 1194#95 | Arena (LMArena) | |
| MMMU-Pro | – | 69.3%#153 | – | Artificial Analysis | |
| VISTA | 51.7%#5 | 51.8%#4 | – | Scale AI SEAL | |
| MMMU (validation)effort not stated | – | – | 81.6%#4 | MMMU | |
| MMMU-Pro (Vals)not in index | – | 79.7%#45 | – | Vals AI | |
| MortgageTaxnot in index | – | 64.8%#45 | – | Vals AI | |
| SAGEnot in index | – | 41.1%#57 | – | Vals AI | |
| VTBnot in indexeffort not stated | – | – | 11.1%#10 | Scale AI SEAL | |
| GeoBenchnot in index | 64.0%#14 | 64.0%#14 | – | GeoBench | |
| VPCTnot in index | 57.5%#8 | – | – | Epoch AI Benchmarking Hub |
Long context
| Benchmark | medium | high | Source | Trend |
|---|---|---|---|---|
| Fiction.LiveBench 120k | 62.5%#13 | – | Fiction.live | |
| AA-LCR | – | 61.0%#278 | Artificial Analysis |
Composite
| Benchmark | high | not stated | Source | Trend |
|---|---|---|---|---|
| Epoch Capabilities Indexnot in indexeffort not stated | – | 145.5#83 | Epoch AI Benchmarking Hub | |
| AA Intelligence Index v4.3.2 | 16.6#270 | – | Artificial Analysis | |
| HELM Capabilities meannot in indexeffort not stated | – | 81.2%#2 | HELM Capabilities |
Safety & honesty
| Benchmark | medium | high | not stated | Source | Trend |
|---|---|---|---|---|---|
| FORTRESSnot in indexeffort not stated | – | – | 21.5%#36 | Scale AI SEAL | |
| MASKnot in indexpossible contamination | 72.9%#28 | 78.6%#27 | – | Scale AI SEAL | |
| PropensityBenchnot in indexeffort not stated | – | – | 15.8%#12 | Scale AI SEAL |
Where it wins
Benchmarks where this configuration ranks in the top five of every configuration measured.
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 $0.275 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0018 | $0.0015 | – |
| Summarize a 30-page report | 12,000 / 600 | $0.016 | $0.0084 | – |
| Code edit | 6,000 / 1,500 | $0.013 | $0.0095 | – |
| Agentic coding session | 60,000 / 4,000 | $0.084 | $0.046 | – |
| Structured extraction | 2,000 / 200 | $0.0031 | $0.0018 | – |
See also
Questions people ask
- What are o4-mini's benchmark scores?
- As of 25 Oct 2026, o4-mini scores 56.9 ±5.9 on the BenchLeader Index, strongest in maths (69) and weakest in knowledge (38). That rests on 15 independent benchmark results, and every score on this page links to its publisher.
- How does o4-mini rank?
- o4-mini is #192 of 785 ranked configurations on 25 Oct 2026. 299 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 o4-mini cost per 1M tokens?
- As of 25 Oct 2026, o4-mini costs $1.10 per 1M input tokens and $4.40 per 1M output tokens, $1.93 blended (three input tokens per output token).
- When was o4-mini released?
- OpenAI released o4-mini on 16 Apr 2025. BenchLeader has results for it at 4 reasoning-effort settings.
Data as of 25 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “o4-mini: benchmarks, pricing, speed and rank”, data as of 25 Oct 2026, https://www.benchleader.com/models/o4-mini.