Model comparison
Gemini 2.5 Pro vs GPT-5.1
GPT-5.1 is the stronger model overall, scoring 49.0 to 45.0 on the Noometry Index.
Last verified . 59 shared benchmarks.
GoogleSummary
- They share 59 benchmarks with published results for both. Gemini 2.5 Pro scores higher in 4 categories and GPT-5.1 in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.1 leads 52.2 to 32.5.
- The biggest single-benchmark swing is ARC-AGI-1: 41% for Gemini 2.5 Pro and 72.8% for GPT-5.1.
- Both cost about the same: $1.25 input and $10 output per million tokens.
- Gemini 2.5 Pro accepts more context: 1.05M tokens versus 400K.
Side by side
| Gemini 2.5 Pro | GPT-5.1 | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 45.0 | 49.0 |
| Released | 2025-03-25 | 2025-11-13 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 400K |
| Max output | 66K | 128K |
| Input $ / M tokens | $1.25 | $1.25 |
| Output $ / M tokens | $10 | $10 |
| Results tracked | 78 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.1 leads
Gemini 2.5 Pro: 42.4 (#101), GPT-5.1: 46.4 (#66)
| Benchmark | Gemini 2.5 Pro | GPT-5.1 |
|---|---|---|
| SWE-bench Verified | 57.6% | 68% |
| SWE-bench Verified (bash only) | 53.6% | 66% |
| LMArena WebDev | 1227 | 1395 |
| SciCode | 42.8% | 43.3% |
| GSO | 3.9% | 13.7% |
| WeirdML | 54% | 60.8% |
| LiveBench Coding | 85.9% | 72.5% |
| LMArena Coding | 1452 | 1454 |
| ALE-Bench | 785.52 | 1,192 |
| Aider Polyglot | 83.1% | — |
| CadEval | 64% | — |
| AlgoTune | 1.51 | — |
Agentic & Tool Use GPT-5.1 leads
Gemini 2.5 Pro: 29.2 (#88), GPT-5.1: 32.7 (#60)
| Benchmark | Gemini 2.5 Pro | GPT-5.1 |
|---|---|---|
| Terminal-Bench | 32.6% | 47.6% |
| DeepResearch Bench | 42.8% | 42.8% |
| LMArena Search | 1142 | 1199 |
| Vending-Bench 2 | 573.64 | 1,473 |
| GDPval | 23.3% | — |
| Remote Labor Index | 0.8% | — |
| TheAgentCompany | 30.3% | — |
| τ²-bench Banking | 13.7% | — |
| BALROG | 43.3% | — |
| METR Time Horizons | 55.4% | — |
Reasoning GPT-5.1 leads
Gemini 2.5 Pro: 28.8 (#99), GPT-5.1: 39.8 (#58)
| Benchmark | Gemini 2.5 Pro | GPT-5.1 |
|---|---|---|
| ARC-AGI-2 | 4.9% | 17.6% |
| SimpleBench | 62.4% | 53.2% |
| ARC-AGI-1 | 41% | 72.8% |
| CritPt | 2% | 4.9% |
| Chess Puzzles | 20% | 32% |
| EnigmaEval | 5.6% | 11.2% |
| LiveBench Reasoning | 89.8% | 95.8% |
| LMArena Hard Prompts | 1455 | 1457 |
| DTBench | 82.4% | 90.1% |
| LiveBench Data Analysis | 79.9% | 72.1% |
| LMCA | 34.8% | 43.9% |
| Epoch Capabilities Index | 145.32 | 149.64 |
| ForecastBench | 61.3 | 58.1 |
| LiveBench | 82.3% | 78.8% |
| Kagi LLM Benchmark | 70.3% | — |
| Mystery Game Puzzles | — | 19% |
Math GPT-5.1 leads
Gemini 2.5 Pro: 32.5 (#213), GPT-5.1: 52.2 (#51)
| Benchmark | Gemini 2.5 Pro | GPT-5.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 84.7% | 88.6% |
| Omni-MATH | 41.6% | 46.4% |
| LiveBench Math | 90.2% | 94.5% |
| LMArena Math | 1450 | 1447 |
| FrontierMath (Feb 2025 set) | 14.1% | 31% |
| FrontierMath Tier 4 (v1) | 4.2% | 12.5% |
| FrontierMath (Tiers 1-3) | 24.6% | — |
| FrontierMath Tier 4 | 0% | — |
| MATH Level 5 | 95.9% | — |
Knowledge Gemini 2.5 Pro leads
Gemini 2.5 Pro: 56.0 (#46), GPT-5.1: 50.6 (#71)
| Benchmark | Gemini 2.5 Pro | GPT-5.1 |
|---|---|---|
| GPQA Diamond | 85.3% | 87.6% |
| Humanity's Last Exam | 21.6% | 23.7% |
| MMLU-Pro | 86.3% | 57.9% |
| Vectara Hallucination Rate | 7% | 10.9% |
| GPQA (HELM) | 74.9% | 44.2% |
| LMArena Expert | 1452 | 1470 |
| SimpleQA Verified | — | 48% |
| Confabulations | 10.6% | — |
Multimodal Too close to call
Gemini 2.5 Pro: 45.2 (#18), GPT-5.1: 44.8 (#19)
| Benchmark | Gemini 2.5 Pro | GPT-5.1 |
|---|---|---|
| LMArena Vision | 1263 | 1250 |
| VPCT | 48% | 58.7% |
| LMArena Document | 1421 | 1403 |
| GeoBench | 86% | — |
| SpatialViz-Bench | 44.7% | — |
Multilingual Gemini 2.5 Pro leads
Gemini 2.5 Pro: 55.3 (#31), GPT-5.1: 53.8 (#56)
| Benchmark | Gemini 2.5 Pro | GPT-5.1 |
|---|---|---|
| LMArena Non-English | 1451 | 1431 |
| LMArena Chinese | 1507 | 1495 |
| LMArena French | 1472 | 1450 |
| LMArena German | 1487 | 1438 |
| LMArena Japanese | 1461 | 1453 |
| LMArena Korean | 1434 | 1401 |
| LMArena Russian | 1461 | 1435 |
| LMArena Spanish | 1473 | 1433 |
Instruction Following GPT-5.1 leads
Gemini 2.5 Pro: 75.0 (#75), GPT-5.1: 83.9 (#1)
| Benchmark | Gemini 2.5 Pro | GPT-5.1 |
|---|---|---|
| LiveBench Instruction Following | 80.6% | 93.3% |
| IFEval | 84% | 93.5% |
| LMArena Instruction Following | 1437 | 1443 |
Long Context Gemini 2.5 Pro leads
Gemini 2.5 Pro: 59.8 (#5), GPT-5.1: 47.6 (#14)
| Benchmark | Gemini 2.5 Pro | GPT-5.1 |
|---|---|---|
| LMArena Longer Query | 1449 | 1447 |
| Fiction.LiveBench | 91.7% | — |
| CL-bench | — | 23.7% |
| CL-bench Life | — | 17.3% |
Writing & Preference Too close to call
Gemini 2.5 Pro: 63.7 (#62), GPT-5.1: 64.5 (#55)
| Benchmark | Gemini 2.5 Pro | GPT-5.1 |
|---|---|---|
| LMArena Text | 1458 | 1443 |
| LMArena Creative Writing | 1454 | 1427 |
| WildBench | 85.7% | 86.3% |
| LMArena Multi-Turn | 1453 | 1450 |
| LiveBench Language | 67.8% | 80.2% |
| Short-Story Creative Writing | 83.8% | — |
| EQ-Bench Creative Writing | 1421 | — |
Frequently asked questions
Is Gemini 2.5 Pro better than GPT-5.1?
GPT-5.1 is the stronger model overall, scoring 49.0 to 45.0 on the Noometry Index.
Which is cheaper, Gemini 2.5 Pro or GPT-5.1?
GPT-5.1 is cheaper. It lists at $1.25 per million input tokens and $10 per million output tokens; Gemini 2.5 Pro lists at $1.25 and $10.
Is Gemini 2.5 Pro or GPT-5.1 better for coding?
GPT-5.1 scores higher on coding benchmarks: 46.4 versus 42.4 in the Noometry coding category.
Which has the bigger context window?
Gemini 2.5 Pro does, with 1.05M tokens against 400K.
How many benchmarks do Gemini 2.5 Pro and GPT-5.1 share?
59 benchmarks have published results for both models. Gemini 2.5 Pro has 78 scored results on Noometry and GPT-5.1 has 63.