Model comparison
GPT-5.1 vs GPT-5.5
GPT-5.5 is the stronger model overall, scoring 63.4 to 49.0 on the Noometry Index. GPT-5.1 costs 3.3× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.
Last verified . 46 shared benchmarks.
Summary
- They share 46 benchmarks with published results for both. GPT-5.1 scores higher in 1 category and GPT-5.5 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.5 leads 72.8 to 39.8.
- The biggest single-benchmark swing is ARC-AGI-2: 17.6% for GPT-5.1 and 85% for GPT-5.5.
- GPT-5.1 is cheaper at $1.25 / $10 per million input/output tokens, against $5 / $30 for GPT-5.5.
- GPT-5.5 accepts more context: 1.05M tokens versus 400K.
Side by side
| GPT-5.1 | GPT-5.5 | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 49.0 | 63.4 |
| Released | 2025-11-13 | 2026-04-23 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 1.05M |
| Max output | 128K | 128K |
| Input $ / M tokens | $1.25 | $5 |
| Output $ / M tokens | $10 | $30 |
| Results tracked | 63 | 71 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.5 leads
GPT-5.1: 46.4 (#66), GPT-5.5: 58.2 (#17)
| Benchmark | GPT-5.1 | GPT-5.5 |
|---|---|---|
| SWE-bench Verified | 68% | 80.6% |
| LMArena WebDev | 1395 | 1513 |
| SciCode | 43.3% | 56.1% |
| GSO | 13.7% | 40.2% |
| WeirdML | 60.8% | 84.9% |
| LMArena Coding | 1454 | 1494 |
| ALE-Bench | 1,192 | 1,943 |
| DeepSWE | — | 67% |
| FrontierCode | — | 43% |
| SWE-bench Verified (bash only) | 66% | — |
| LiveBench Coding | 72.5% | — |
| MirrorCode | — | 10% |
Agentic & Tool Use GPT-5.5 leads
GPT-5.1: 32.7 (#60), GPT-5.5: 50.7 (#6)
| Benchmark | GPT-5.1 | GPT-5.5 |
|---|---|---|
| Terminal-Bench | 47.6% | 84.7% |
| DeepResearch Bench | 42.8% | 54% |
| LMArena Search | 1199 | 1242 |
| Vending-Bench 2 | 1,473 | 7,524 |
| APEX-Agents | — | 55.1% |
| OSWorld 2.0 | — | 13% |
| Remote Labor Index | — | 6.3% |
| τ²-bench Banking | — | 44.6% |
| PostTrainBench | — | 27.2% |
| ExploitBench | — | 47.4% |
| GBAEval | — | 53.2% |
| GDP.pdf | — | 26% |
Reasoning GPT-5.5 leads
GPT-5.1: 39.8 (#58), GPT-5.5: 72.8 (#11)
| Benchmark | GPT-5.1 | GPT-5.5 |
|---|---|---|
| ARC-AGI-2 | 17.6% | 85% |
| SimpleBench | 53.2% | 69% |
| ARC-AGI-1 | 72.8% | 95% |
| CritPt | 4.9% | 27.1% |
| Chess Puzzles | 32% | 54% |
| LMArena Hard Prompts | 1457 | 1489 |
| Mystery Game Puzzles | 19% | 56% |
| DTBench | 90.1% | 96% |
| LMCA | 43.9% | 54.3% |
| Epoch Capabilities Index | 149.64 | 159.1 |
| ForecastBench | 58.1 | 60.6 |
| Kagi LLM Benchmark | — | 88.8% |
| NYT Connections (extended) | — | 96.2% |
| EnigmaEval | 11.2% | — |
| EBR-Bench | — | 34.3% |
| LiveBench Reasoning | 95.8% | — |
| LiveBench Data Analysis | 72.1% | — |
| Surface Evolver Bench | — | 88.1% |
| Bench to the Future 3 | — | 0.14 |
| LiveBench | 78.8% | — |
Math GPT-5.5 leads
GPT-5.1: 52.2 (#51), GPT-5.5: 81.7 (#11)
| Benchmark | GPT-5.1 | GPT-5.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.6% | 100% |
| LMArena Math | 1447 | 1486 |
| FrontierMath (Feb 2025 set) | 31% | 51.7% |
| FrontierMath Tier 4 (v1) | 12.5% | 35.4% |
| FrontierMath (Tiers 1-3) | — | 85.3% |
| FrontierMath Tier 4 | — | 72.5% |
| MathArena Final-Answer Competitions | — | 94.3% |
| ProofBench | — | 50% |
| Omni-MATH | 46.4% | — |
| LiveBench Math | 94.5% | — |
| FrontierMath Erdős | — | 0% |
Knowledge GPT-5.5 leads
GPT-5.1: 50.6 (#71), GPT-5.5: 64.4 (#17)
| Benchmark | GPT-5.1 | GPT-5.5 |
|---|---|---|
| GPQA Diamond | 87.6% | 94% |
| SimpleQA Verified | 48% | 63% |
| Vectara Hallucination Rate | 10.9% | 9.3% |
| LMArena Expert | 1470 | 1508 |
| Humanity's Last Exam | 23.7% | — |
| MMLU-Pro | 57.9% | — |
| GPQA (HELM) | 44.2% | — |
Multimodal GPT-5.5 leads
GPT-5.1: 44.8 (#19), GPT-5.5: 46.9 (#12)
| Benchmark | GPT-5.1 | GPT-5.5 |
|---|---|---|
| LMArena Vision | 1250 | 1297 |
| LMArena Document | 1403 | 1486 |
| VPCT | 58.7% | — |
| Blueprint-Bench 2 | — | 36.2% |
| Furniture Assembly | — | 44.2% |
Multilingual GPT-5.5 leads
GPT-5.1: 53.8 (#56), GPT-5.5: 56.4 (#20)
| Benchmark | GPT-5.1 | GPT-5.5 |
|---|---|---|
| LMArena Non-English | 1431 | 1467 |
| LMArena Chinese | 1495 | 1533 |
| LMArena French | 1450 | 1486 |
| LMArena German | 1438 | 1480 |
| LMArena Japanese | 1453 | 1498 |
| LMArena Korean | 1401 | 1460 |
| LMArena Russian | 1435 | 1473 |
| LMArena Spanish | 1433 | 1468 |
Instruction Following GPT-5.1 leads
GPT-5.1: 83.9 (#1), GPT-5.5: 77.5 (#18)
| Benchmark | GPT-5.1 | GPT-5.5 |
|---|---|---|
| LMArena Instruction Following | 1443 | 1479 |
| LiveBench Instruction Following | 93.3% | — |
| IFEval | 93.5% | — |
Long Context Too close to call
GPT-5.1: 47.6 (#14), GPT-5.5: 48.3 (#12)
| Benchmark | GPT-5.1 | GPT-5.5 |
|---|---|---|
| CL-bench Life | 17.3% | 22.2% |
| LMArena Longer Query | 1447 | 1484 |
| CL-bench | 23.7% | — |
Writing & Preference GPT-5.5 leads
GPT-5.1: 64.5 (#55), GPT-5.5: 72.7 (#13)
| Benchmark | GPT-5.1 | GPT-5.5 |
|---|---|---|
| LMArena Text | 1443 | 1472 |
| LMArena Creative Writing | 1427 | 1455 |
| LMArena Multi-Turn | 1450 | 1476 |
| EQ-Bench Creative Writing | — | 1844 |
| WildBench | 86.3% | — |
| EQ-Bench 4 | — | 1315 |
| LiveBench Language | 80.2% | — |
Frequently asked questions
Is GPT-5.1 better than GPT-5.5?
GPT-5.5 is the stronger model overall, scoring 63.4 to 49.0 on the Noometry Index. GPT-5.1 costs 3.3× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.
Which is cheaper, GPT-5.1 or GPT-5.5?
GPT-5.1 is cheaper. It lists at $1.25 per million input tokens and $10 per million output tokens; GPT-5.5 lists at $5 and $30.
Is GPT-5.1 or GPT-5.5 better for coding?
GPT-5.5 scores higher on coding benchmarks: 58.2 versus 46.4 in the Noometry coding category.
Which has the bigger context window?
GPT-5.5 does, with 1.05M tokens against 400K.
How many benchmarks do GPT-5.1 and GPT-5.5 share?
46 benchmarks have published results for both models. GPT-5.1 has 63 scored results on Noometry and GPT-5.5 has 71.