Model comparison
GPT-5.1 vs o1
GPT-5.1 is the stronger model overall, scoring 49.0 to 40.9 on the Noometry Index.
Last verified . 39 shared benchmarks.
Summary
- They share 39 benchmarks with published results for both. GPT-5.1 scores higher in 9 categories and o1 in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.1 leads 52.2 to 36.1.
- The biggest single-benchmark swing is ARC-AGI-1: 72.8% for GPT-5.1 and 30.7% for o1.
- GPT-5.1 is cheaper at $1.25 / $10 per million input/output tokens, against $15 / $60 for o1.
- GPT-5.1 accepts more context: 400K tokens versus 200K.
Side by side
| GPT-5.1 | o1 | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 49.0 | 40.9 |
| Released | 2025-11-13 | 2024-09-12 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 200K |
| Max output | 128K | 100K |
| Input $ / M tokens | $1.25 | $15 |
| Output $ / M tokens | $10 | $60 |
| Results tracked | 63 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
GPT-5.1: 46.4 (#66), o1: 46.1 (#70)
| Benchmark | GPT-5.1 | o1 |
|---|---|---|
| WeirdML | 60.8% | 47.6% |
| LiveBench Coding | 72.5% | 69.7% |
| LMArena Coding | 1454 | 1367 |
| SWE-bench Verified | 68% | — |
| SWE-bench Verified (bash only) | 66% | — |
| Aider Polyglot | — | 61.7% |
| LMArena WebDev | 1395 | — |
| SciCode | 43.3% | — |
| GSO | 13.7% | — |
| CadEval | — | 56% |
| ALE-Bench | 1,192 | — |
| HumanEval+ | — | 89% |
| MBPP+ | — | 80.2% |
Agentic & Tool Use GPT-5.1 leads
GPT-5.1: 32.7 (#60), o1: 24.6 (#117)
| Benchmark | GPT-5.1 | o1 |
|---|---|---|
| Terminal-Bench | 47.6% | — |
| Cybench | — | 10% |
| DeepResearch Bench | 42.8% | — |
| LMArena Search | 1199 | — |
| METR Time Horizons | — | 51.1% |
| Vending-Bench 2 | 1,473 | — |
Reasoning GPT-5.1 leads
GPT-5.1: 39.8 (#58), o1: 27.9 (#111)
| Benchmark | GPT-5.1 | o1 |
|---|---|---|
| SimpleBench | 53.2% | 41.7% |
| ARC-AGI-1 | 72.8% | 30.7% |
| Chess Puzzles | 32% | 15% |
| EnigmaEval | 11.2% | 5.7% |
| LiveBench Reasoning | 95.8% | 91.6% |
| LMArena Hard Prompts | 1457 | 1371 |
| DTBench | 90.1% | 74.7% |
| LiveBench Data Analysis | 72.1% | 65.5% |
| LMCA | 43.9% | 22.3% |
| Epoch Capabilities Index | 149.64 | 141.91 |
| LiveBench | 78.8% | 75.7% |
| ARC-AGI-2 | 17.6% | — |
| CritPt | 4.9% | — |
| Mystery Game Puzzles | 19% | — |
| ForecastBench | 58.1 | — |
Math GPT-5.1 leads
GPT-5.1: 52.2 (#51), o1: 36.1 (#175)
| Benchmark | GPT-5.1 | o1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.6% | 73.3% |
| LiveBench Math | 94.5% | 80.3% |
| LMArena Math | 1447 | 1388 |
| FrontierMath (Feb 2025 set) | 31% | 9.3% |
| FrontierMath (Tiers 1-3) | — | 14.7% |
| Omni-MATH | 46.4% | — |
| MATH Level 5 | — | 94.7% |
| FrontierMath Tier 4 (v1) | 12.5% | — |
Knowledge GPT-5.1 leads
GPT-5.1: 50.6 (#71), o1: 41.5 (#110)
| Benchmark | GPT-5.1 | o1 |
|---|---|---|
| GPQA Diamond | 87.6% | 76.8% |
| Humanity's Last Exam | 23.7% | 8% |
| SimpleQA Verified | 48% | 41.1% |
| LMArena Expert | 1470 | 1361 |
| MMLU-Pro | 57.9% | — |
| Confabulations | — | 11.7% |
| Vectara Hallucination Rate | 10.9% | — |
| GPQA (HELM) | 44.2% | — |
Multimodal GPT-5.1 leads
GPT-5.1: 44.8 (#19), o1: 34.2 (#93)
| Benchmark | GPT-5.1 | o1 |
|---|---|---|
| LMArena Vision | 1250 | 1168 |
| VPCT | 58.7% | 37% |
| GeoBench | — | 80% |
| LMArena Document | 1403 | — |
| SpatialViz-Bench | — | 41.4% |
Multilingual GPT-5.1 leads
GPT-5.1: 53.8 (#56), o1: 48.6 (#142)
| Benchmark | GPT-5.1 | o1 |
|---|---|---|
| LMArena Non-English | 1431 | 1358 |
| LMArena Chinese | 1495 | 1394 |
| LMArena French | 1450 | 1344 |
| LMArena German | 1438 | 1337 |
| LMArena Japanese | 1453 | 1346 |
| LMArena Korean | 1401 | 1396 |
| LMArena Russian | 1435 | 1356 |
| LMArena Spanish | 1433 | 1345 |
Instruction Following GPT-5.1 leads
GPT-5.1: 83.9 (#1), o1: 74.8 (#86)
| Benchmark | GPT-5.1 | o1 |
|---|---|---|
| LiveBench Instruction Following | 93.3% | 81.5% |
| LMArena Instruction Following | 1443 | 1367 |
| IFEval | 93.5% | — |
Long Context o1 leads
GPT-5.1: 47.6 (#14), o1: 50.3 (#9)
| Benchmark | GPT-5.1 | o1 |
|---|---|---|
| LMArena Longer Query | 1447 | 1378 |
| Fiction.LiveBench | — | 83.3% |
| CL-bench | 23.7% | — |
| CL-bench Life | 17.3% | — |
Writing & Preference GPT-5.1 leads
GPT-5.1: 64.5 (#55), o1: 55.6 (#144)
| Benchmark | GPT-5.1 | o1 |
|---|---|---|
| LMArena Text | 1443 | 1366 |
| LMArena Creative Writing | 1427 | 1348 |
| LMArena Multi-Turn | 1450 | 1369 |
| LiveBench Language | 80.2% | 65.4% |
| Short-Story Creative Writing | — | 70.2% |
| WildBench | 86.3% | — |
Frequently asked questions
Is GPT-5.1 better than o1?
GPT-5.1 is the stronger model overall, scoring 49.0 to 40.9 on the Noometry Index.
Which is cheaper, GPT-5.1 or o1?
GPT-5.1 is cheaper. It lists at $1.25 per million input tokens and $10 per million output tokens; o1 lists at $15 and $60.
Is GPT-5.1 or o1 better for coding?
They score almost the same on coding (46.4 vs 46.1); test both on your own repository before choosing.
Which has the bigger context window?
GPT-5.1 does, with 400K tokens against 200K.
How many benchmarks do GPT-5.1 and o1 share?
39 benchmarks have published results for both models. GPT-5.1 has 63 scored results on Noometry and o1 has 52.