Model comparison
GPT-5 Mini vs o1
GPT-5 Mini and o1 score almost the same on the Noometry Index (41.8 vs 40.9), so choose on price, context window or the category you care about most.
Last verified . 36 shared benchmarks.
Summary
- They share 36 benchmarks with published results for both. GPT-5 Mini scores higher in 6 categories and o1 in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Mini leads 46.7 to 36.1.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 46.7% for GPT-5 Mini and 14.7% for o1.
- GPT-5 Mini is cheaper at $0.25 / $2 per million input/output tokens, against $15 / $60 for o1.
- GPT-5 Mini accepts more context: 400K tokens versus 200K.
Side by side
| GPT-5 Mini | o1 | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 41.8 | 40.9 |
| Released | 2025-08-07 | 2024-09-12 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 200K |
| Max output | 128K | 100K |
| Input $ / M tokens | $0.25 | $15 |
| Output $ / M tokens | $2 | $60 |
| Results tracked | 60 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o1 leads
GPT-5 Mini: 40.1 (#146), o1: 46.1 (#70)
| Benchmark | GPT-5 Mini | o1 |
|---|---|---|
| WeirdML | 52.7% | 47.6% |
| LMArena Coding | 1406 | 1367 |
| SWE-bench Verified | 64.7% | — |
| SWE-bench Verified (bash only) | 59.8% | — |
| Aider Polyglot | — | 61.7% |
| SWE-bench Multilingual | 39.7% | — |
| SciCode | 39.2% | — |
| LiveBench Coding | — | 69.7% |
| CadEval | — | 56% |
| ALE-Bench | 799.77 | — |
| AlgoTune | 1.38 | — |
| HumanEval+ | — | 89% |
| MBPP+ | — | 80.2% |
Agentic & Tool Use GPT-5 Mini leads
GPT-5 Mini: 31.1 (#70), o1: 24.6 (#117)
| Benchmark | GPT-5 Mini | o1 |
|---|---|---|
| Terminal-Bench | 34.8% | — |
| Berkeley Function Calling Leaderboard | 55.5% | — |
| Cybench | — | 10% |
| METR Time Horizons | — | 51.1% |
| Vending-Bench 2 | -31.18 | — |
Reasoning o1 leads
GPT-5 Mini: 23.9 (#168), o1: 27.9 (#111)
| Benchmark | GPT-5 Mini | o1 |
|---|---|---|
| ARC-AGI-1 | 54.3% | 30.7% |
| Chess Puzzles | 30% | 15% |
| EnigmaEval | 8.2% | 5.7% |
| LMArena Hard Prompts | 1380 | 1371 |
| DTBench | 80.5% | 74.7% |
| LMCA | 34.2% | 22.3% |
| Epoch Capabilities Index | 145.52 | 141.91 |
| ARC-AGI-2 | 4.4% | — |
| SimpleBench | — | 41.7% |
| Kagi LLM Benchmark | 70.3% | — |
| CritPt | 0% | — |
| LiveBench Reasoning | — | 91.6% |
| Mystery Game Puzzles | 10% | — |
| LiveBench Data Analysis | — | 65.5% |
| ForecastBench | 61 | — |
| LiveBench | — | 75.7% |
Math GPT-5 Mini leads
GPT-5 Mini: 46.7 (#69), o1: 36.1 (#175)
| Benchmark | GPT-5 Mini | o1 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 46.7% | 14.7% |
| OTIS Mock AIME 2024-2025 | 86.7% | 73.3% |
| LMArena Math | 1378 | 1388 |
| MATH Level 5 | 97.8% | 94.7% |
| FrontierMath (Feb 2025 set) | 27.2% | 9.3% |
| FrontierMath Tier 4 | 12.2% | — |
| ProofBench | 9% | — |
| Omni-MATH | 72.2% | — |
| LiveBench Math | — | 80.3% |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge GPT-5 Mini leads
GPT-5 Mini: 45.6 (#86), o1: 41.5 (#110)
| Benchmark | GPT-5 Mini | o1 |
|---|---|---|
| GPQA Diamond | 75% | 76.8% |
| Humanity's Last Exam | 19.4% | 8% |
| SimpleQA Verified | 21.6% | 41.1% |
| Confabulations | 13.3% | 11.7% |
| LMArena Expert | 1379 | 1361 |
| MMLU-Pro | 83.5% | — |
| Vectara Hallucination Rate | 12.9% | — |
| GPQA (HELM) | 75.6% | — |
Multimodal GPT-5 Mini leads
GPT-5 Mini: 35.6 (#85), o1: 34.2 (#93)
| Benchmark | GPT-5 Mini | o1 |
|---|---|---|
| LMArena Vision | 1202 | 1168 |
| VPCT | 40.2% | 37% |
| GeoBench | — | 80% |
| SpatialViz-Bench | — | 41.4% |
Multilingual Too close to call
GPT-5 Mini: 48.9 (#137), o1: 48.6 (#142)
| Benchmark | GPT-5 Mini | o1 |
|---|---|---|
| LMArena Non-English | 1363 | 1358 |
| LMArena Chinese | 1385 | 1394 |
| LMArena French | 1386 | 1344 |
| LMArena German | 1366 | 1337 |
| LMArena Japanese | 1341 | 1346 |
| LMArena Korean | 1308 | 1396 |
| LMArena Russian | 1362 | 1356 |
| LMArena Spanish | 1355 | 1345 |
Instruction Following GPT-5 Mini leads
GPT-5 Mini: 76.2 (#46), o1: 74.8 (#86)
| Benchmark | GPT-5 Mini | o1 |
|---|---|---|
| LMArena Instruction Following | 1357 | 1367 |
| LiveBench Instruction Following | — | 81.5% |
| IFEval | 92.7% | — |
Long Context o1 leads
GPT-5 Mini: 41.9 (#132), o1: 50.3 (#9)
| Benchmark | GPT-5 Mini | o1 |
|---|---|---|
| Fiction.LiveBench | 69.4% | 83.3% |
| LMArena Longer Query | 1355 | 1378 |
Writing & Preference Too close to call
GPT-5 Mini: 55.2 (#148), o1: 55.6 (#144)
| Benchmark | GPT-5 Mini | o1 |
|---|---|---|
| LMArena Text | 1373 | 1366 |
| LMArena Creative Writing | 1325 | 1348 |
| Short-Story Creative Writing | 83.1% | 70.2% |
| LMArena Multi-Turn | 1363 | 1369 |
| EQ-Bench Creative Writing | 1313 | — |
| WildBench | 85.5% | — |
| LiveBench Language | — | 65.4% |
Frequently asked questions
Is GPT-5 Mini better than o1?
GPT-5 Mini and o1 score almost the same on the Noometry Index (41.8 vs 40.9), so choose on price, context window or the category you care about most.
Which is cheaper, GPT-5 Mini or o1?
GPT-5 Mini is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; o1 lists at $15 and $60.
Is GPT-5 Mini or o1 better for coding?
o1 scores higher on coding benchmarks: 46.1 versus 40.1 in the Noometry coding category.
Which has the bigger context window?
GPT-5 Mini does, with 400K tokens against 200K.
How many benchmarks do GPT-5 Mini and o1 share?
36 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and o1 has 52.