Model comparison
Claude 3.5 Haiku vs GPT-5.1
GPT-5.1 is the stronger model overall, scoring 49.0 to 29.2 on the Noometry Index.
Last verified . 38 shared benchmarks.
Summary
- They share 38 benchmarks with published results for both. Claude 3.5 Haiku scores higher in 0 categories and GPT-5.1 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.1 leads 52.2 to 14.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.3% for Claude 3.5 Haiku and 88.6% for GPT-5.1.
Side by side
| Claude 3.5 Haiku | GPT-5.1 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 29.2 | 49.0 |
| Released | 2024-10-22 | 2025-11-13 |
| Weights | Proprietary | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $1.25 |
| Output $ / M tokens | — | $10 |
| Results tracked | 49 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.1 leads
Claude 3.5 Haiku: 32.9 (#265), GPT-5.1: 46.4 (#66)
| Benchmark | Claude 3.5 Haiku | GPT-5.1 |
|---|---|---|
| SciCode | 27.4% | 43.3% |
| WeirdML | 30.7% | 60.8% |
| LiveBench Coding | 51.4% | 72.5% |
| LMArena Coding | 1286 | 1454 |
| SWE-bench Verified | — | 68% |
| SWE-bench Verified (bash only) | — | 66% |
| Aider Polyglot | 28% | — |
| LMArena WebDev | — | 1395 |
| GSO | — | 13.7% |
| BigCodeBench Instruct | 46.1% | — |
| BigCodeBench Complete | 59% | — |
| CadEval | 32% | — |
| ALE-Bench | — | 1,192 |
Agentic & Tool Use GPT-5.1 leads
Claude 3.5 Haiku: 28.0 (#95), GPT-5.1: 32.7 (#60)
| Benchmark | Claude 3.5 Haiku | GPT-5.1 |
|---|---|---|
| Terminal-Bench | — | 47.6% |
| DeepResearch Bench | — | 42.8% |
| BALROG | 19.3% | — |
| LMArena Search | — | 1199 |
| Vending-Bench 2 | — | 1,473 |
Reasoning GPT-5.1 leads
Claude 3.5 Haiku: 17.7 (#290), GPT-5.1: 39.8 (#58)
| Benchmark | Claude 3.5 Haiku | GPT-5.1 |
|---|---|---|
| CritPt | 0% | 4.9% |
| LiveBench Reasoning | 28.1% | 95.8% |
| LMArena Hard Prompts | 1251 | 1457 |
| DTBench | 56.7% | 90.1% |
| LiveBench Data Analysis | 48.5% | 72.1% |
| Epoch Capabilities Index | 127.15 | 149.64 |
| LiveBench | 43.5% | 78.8% |
| ARC-AGI-2 | — | 17.6% |
| SimpleBench | — | 53.2% |
| ARC-AGI-1 | — | 72.8% |
| Chess Puzzles | — | 32% |
| EnigmaEval | — | 11.2% |
| Mystery Game Puzzles | — | 19% |
| LMCA | — | 43.9% |
| ForecastBench | — | 58.1 |
Math GPT-5.1 leads
Claude 3.5 Haiku: 14.7 (#300), GPT-5.1: 52.2 (#51)
| Benchmark | Claude 3.5 Haiku | GPT-5.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.3% | 88.6% |
| Omni-MATH | 22.4% | 46.4% |
| LiveBench Math | 35.5% | 94.5% |
| LMArena Math | 1244 | 1447 |
| FrontierMath (Feb 2025 set) | 0.3% | 31% |
| MATH Level 5 | 46.4% | — |
| FrontierMath Tier 4 (v1) | — | 12.5% |
Knowledge GPT-5.1 leads
Claude 3.5 Haiku: 18.7 (#281), GPT-5.1: 50.6 (#71)
| Benchmark | Claude 3.5 Haiku | GPT-5.1 |
|---|---|---|
| GPQA Diamond | 38.1% | 87.6% |
| MMLU-Pro | 60.5% | 57.9% |
| GPQA (HELM) | 36.3% | 44.2% |
| LMArena Expert | 1208 | 1470 |
| Humanity's Last Exam | — | 23.7% |
| SimpleQA Verified | — | 48% |
| Confabulations | 36.7% | — |
| Vectara Hallucination Rate | — | 10.9% |
| MMLU | 74.3% | — |
Multimodal GPT-5.1 leads
Claude 3.5 Haiku: 26.8 (#117), GPT-5.1: 44.8 (#19)
| Benchmark | Claude 3.5 Haiku | GPT-5.1 |
|---|---|---|
| LMArena Vision | 1092 | 1250 |
| GeoBench | 34% | — |
| VPCT | — | 58.7% |
| LMArena Document | — | 1403 |
Multilingual GPT-5.1 leads
Claude 3.5 Haiku: 40.0 (#218), GPT-5.1: 53.8 (#56)
| Benchmark | Claude 3.5 Haiku | GPT-5.1 |
|---|---|---|
| LMArena Non-English | 1238 | 1431 |
| LMArena Chinese | 1229 | 1495 |
| LMArena French | 1264 | 1450 |
| LMArena German | 1237 | 1438 |
| LMArena Japanese | 1175 | 1453 |
| LMArena Korean | 1173 | 1401 |
| LMArena Russian | 1253 | 1435 |
| LMArena Spanish | 1261 | 1433 |
Instruction Following GPT-5.1 leads
Claude 3.5 Haiku: 62.9 (#234), GPT-5.1: 83.9 (#1)
| Benchmark | Claude 3.5 Haiku | GPT-5.1 |
|---|---|---|
| LiveBench Instruction Following | 61.9% | 93.3% |
| IFEval | 79.2% | 93.5% |
| LMArena Instruction Following | 1241 | 1443 |
Long Context GPT-5.1 leads
Claude 3.5 Haiku: 38.3 (#200), GPT-5.1: 47.6 (#14)
| Benchmark | Claude 3.5 Haiku | GPT-5.1 |
|---|---|---|
| LMArena Longer Query | 1261 | 1447 |
| CL-bench | — | 23.7% |
| CL-bench Life | — | 17.3% |
Writing & Preference GPT-5.1 leads
Claude 3.5 Haiku: 42.7 (#234), GPT-5.1: 64.5 (#55)
| Benchmark | Claude 3.5 Haiku | GPT-5.1 |
|---|---|---|
| LMArena Text | 1255 | 1443 |
| LMArena Creative Writing | 1233 | 1427 |
| WildBench | 76% | 86.3% |
| LMArena Multi-Turn | 1265 | 1450 |
| LiveBench Language | 35.4% | 80.2% |
| Short-Story Creative Writing | 73.5% | — |
| EQ-Bench Creative Writing | 1146 | — |
Frequently asked questions
Is Claude 3.5 Haiku better than GPT-5.1?
GPT-5.1 is the stronger model overall, scoring 49.0 to 29.2 on the Noometry Index.
Is Claude 3.5 Haiku or GPT-5.1 better for coding?
GPT-5.1 scores higher on coding benchmarks: 46.4 versus 32.9 in the Noometry coding category.
How many benchmarks do Claude 3.5 Haiku and GPT-5.1 share?
38 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and GPT-5.1 has 63.