Model comparison
GPT-5 Mini vs o3-mini
GPT-5 Mini is the stronger model overall, scoring 41.8 to 36.7 on the Noometry Index.
Last verified . 39 shared benchmarks.
Summary
- They share 39 benchmarks with published results for both. GPT-5 Mini scores higher in 8 categories and o3-mini in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Mini leads 46.7 to 28.1.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 46.7% for GPT-5 Mini and 18.6% for o3-mini.
- GPT-5 Mini is cheaper at $0.25 / $2 per million input/output tokens, against $1.10 / $4.40 for o3-mini.
- GPT-5 Mini accepts more context: 400K tokens versus 200K.
Side by side
| GPT-5 Mini | o3-mini | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 41.8 | 36.7 |
| Released | 2025-08-07 | 2024-12-20 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 200K |
| Max output | 128K | 100K |
| Input $ / M tokens | $0.25 | $1.10 |
| Output $ / M tokens | $2 | $4.40 |
| Results tracked | 60 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
GPT-5 Mini: 40.1 (#146), o3-mini: 40.8 (#132)
| Benchmark | GPT-5 Mini | o3-mini |
|---|---|---|
| SciCode | 39.2% | 39.8% |
| WeirdML | 52.7% | 43.7% |
| LMArena Coding | 1406 | 1378 |
| SWE-bench Verified | 64.7% | — |
| SWE-bench Verified (bash only) | 59.8% | — |
| Aider Polyglot | — | 60.4% |
| SWE-bench Multilingual | 39.7% | — |
| GSO | — | 1.3% |
| LiveBench Coding | — | 82.7% |
| CadEval | — | 54% |
| ALE-Bench | 799.77 | — |
| AlgoTune | 1.38 | — |
Agentic & Tool Use GPT-5 Mini leads
GPT-5 Mini: 31.1 (#70), o3-mini: 29.6 (#84)
| Benchmark | GPT-5 Mini | o3-mini |
|---|---|---|
| Terminal-Bench | 34.8% | — |
| Berkeley Function Calling Leaderboard | 55.5% | — |
| Cybench | — | 22.5% |
| Vending-Bench 2 | -31.18 | — |
Reasoning GPT-5 Mini leads
GPT-5 Mini: 23.9 (#168), o3-mini: 16.3 (#305)
| Benchmark | GPT-5 Mini | o3-mini |
|---|---|---|
| ARC-AGI-2 | 4.4% | 3% |
| ARC-AGI-1 | 54.3% | 34.5% |
| CritPt | 0% | 0.3% |
| Chess Puzzles | 30% | 17% |
| LMArena Hard Prompts | 1380 | 1366 |
| Mystery Game Puzzles | 10% | 7% |
| DTBench | 80.5% | 68.8% |
| LMCA | 34.2% | 19% |
| Epoch Capabilities Index | 145.52 | 140.34 |
| ForecastBench | 61 | 59.6 |
| SimpleBench | — | 22.8% |
| Kagi LLM Benchmark | 70.3% | — |
| EnigmaEval | 8.2% | — |
| LiveBench Reasoning | — | 89.6% |
| LiveBench Data Analysis | — | 70.6% |
| LiveBench | — | 75.9% |
Math GPT-5 Mini leads
GPT-5 Mini: 46.7 (#69), o3-mini: 28.1 (#244)
| Benchmark | GPT-5 Mini | o3-mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 46.7% | 18.6% |
| FrontierMath Tier 4 | 12.2% | 0% |
| OTIS Mock AIME 2024-2025 | 86.7% | 76.9% |
| LMArena Math | 1378 | 1396 |
| MATH Level 5 | 97.8% | 96.5% |
| FrontierMath (Feb 2025 set) | 27.2% | 12.4% |
| FrontierMath Tier 4 (v1) | 6.3% | 4.2% |
| ProofBench | 9% | — |
| Omni-MATH | 72.2% | — |
| LiveBench Math | — | 77.3% |
Knowledge GPT-5 Mini leads
GPT-5 Mini: 45.6 (#86), o3-mini: 38.3 (#146)
| Benchmark | GPT-5 Mini | o3-mini |
|---|---|---|
| GPQA Diamond | 75% | 77% |
| SimpleQA Verified | 21.6% | 15.3% |
| Confabulations | 13.3% | 17.9% |
| LMArena Expert | 1379 | 1364 |
| Humanity's Last Exam | 19.4% | — |
| MMLU-Pro | 83.5% | — |
| Vectara Hallucination Rate | 12.9% | — |
| GPQA (HELM) | 75.6% | — |
Multimodal Not comparable
GPT-5 Mini: 35.6 (#85), o3-mini: —
| Benchmark | GPT-5 Mini | o3-mini |
|---|---|---|
| LMArena Vision | 1202 | — |
| VPCT | 40.2% | — |
Multilingual GPT-5 Mini leads
GPT-5 Mini: 48.9 (#137), o3-mini: 45.7 (#164)
| Benchmark | GPT-5 Mini | o3-mini |
|---|---|---|
| LMArena Non-English | 1363 | 1319 |
| LMArena Chinese | 1385 | 1379 |
| LMArena French | 1386 | 1334 |
| LMArena German | 1366 | 1303 |
| LMArena Japanese | 1341 | 1286 |
| LMArena Korean | 1308 | 1314 |
| LMArena Russian | 1362 | 1304 |
| LMArena Spanish | 1355 | 1321 |
Instruction Following GPT-5 Mini leads
GPT-5 Mini: 76.2 (#46), o3-mini: 75.1 (#72)
| Benchmark | GPT-5 Mini | o3-mini |
|---|---|---|
| LMArena Instruction Following | 1357 | 1337 |
| LiveBench Instruction Following | — | 84.4% |
| IFEval | 92.7% | — |
Long Context GPT-5 Mini leads
GPT-5 Mini: 41.9 (#132), o3-mini: 33.8 (#256)
| Benchmark | GPT-5 Mini | o3-mini |
|---|---|---|
| Fiction.LiveBench | 69.4% | 50% |
| LMArena Longer Query | 1355 | 1343 |
Writing & Preference GPT-5 Mini leads
GPT-5 Mini: 55.2 (#148), o3-mini: 50.3 (#182)
| Benchmark | GPT-5 Mini | o3-mini |
|---|---|---|
| LMArena Text | 1373 | 1337 |
| LMArena Creative Writing | 1325 | 1286 |
| Short-Story Creative Writing | 83.1% | 61.7% |
| LMArena Multi-Turn | 1363 | 1320 |
| EQ-Bench Creative Writing | 1313 | — |
| WildBench | 85.5% | — |
| LiveBench Language | — | 50.7% |
Frequently asked questions
Is GPT-5 Mini better than o3-mini?
GPT-5 Mini is the stronger model overall, scoring 41.8 to 36.7 on the Noometry Index.
Which is cheaper, GPT-5 Mini or o3-mini?
GPT-5 Mini is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; o3-mini lists at $1.10 and $4.40.
Is GPT-5 Mini or o3-mini better for coding?
They score almost the same on coding (40.1 vs 40.8); test both on your own repository before choosing.
Which has the bigger context window?
GPT-5 Mini does, with 400K tokens against 200K.
How many benchmarks do GPT-5 Mini and o3-mini share?
39 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and o3-mini has 51.