Model comparison
Claude 3.5 Haiku vs GPT-5
GPT-5 is the stronger model overall, scoring 50.9 to 28.9 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 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 leads 54.6 to 13.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.3% for Claude 3.5 Haiku and 91.4% for GPT-5.
Side by side
| Claude 3.5 Haiku | GPT-5 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 28.9 | 50.9 |
| Released | 2024-10-22 | 2025-08-07 |
| Weights | Proprietary | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $1.25 |
| Output $ / M tokens | — | $10 |
| Results tracked | 49 | 69 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 leads
Claude 3.5 Haiku: 32.6 (#258), GPT-5: 50.7 (#39)
| Benchmark | Claude 3.5 Haiku | GPT-5 |
|---|---|---|
| Aider Polyglot | 28% | 88% |
| SciCode | 27.4% | 42.9% |
| WeirdML | 30.7% | 60.7% |
| LMArena Coding | 1287 | 1436 |
| SWE-bench Verified | — | 73.6% |
| SWE-bench Verified (bash only) | — | 65% |
| LMArena WebDev | — | 1417 |
| GSO | — | 6.9% |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | 51.4% | — |
| BigCodeBench Complete | 59% | — |
| CadEval | 32% | — |
| ALE-Bench | — | 1,162 |
| AlgoTune | — | 1.67 |
Agentic & Tool Use GPT-5 leads
Claude 3.5 Haiku: 27.7 (#90), GPT-5: 33.0 (#53)
| Benchmark | Claude 3.5 Haiku | GPT-5 |
|---|---|---|
| BALROG | 19.3% | 32.8% |
| Terminal-Bench | — | 49.6% |
| GDPval | — | 34.8% |
| Remote Labor Index | — | 1.7% |
| DeepResearch Bench | — | 49.6% |
| LMArena Search | — | 1133 |
| METR Time Horizons | — | 69.6% |
Reasoning GPT-5 leads
Claude 3.5 Haiku: 17.1 (#286), GPT-5: 38.3 (#55)
| Benchmark | Claude 3.5 Haiku | GPT-5 |
|---|---|---|
| CritPt | 0% | 12.6% |
| LMArena Hard Prompts | 1251 | 1415 |
| DTBench | 56.7% | 90.7% |
| Epoch Capabilities Index | 127.15 | 150 |
| ARC-AGI-2 | — | 9.9% |
| SimpleBench | — | 56.7% |
| Kagi LLM Benchmark | — | 72.7% |
| ARC-AGI-1 | — | 65.7% |
| Chess Puzzles | — | 37% |
| EnigmaEval | — | 10.5% |
| EBR-Bench | — | 12.7% |
| LiveBench Reasoning | 28.1% | — |
| Mystery Game Puzzles | — | 23% |
| LiveBench Data Analysis | 48.5% | — |
| LMCA | — | 40% |
| ForecastBench | — | 61.4 |
| LiveBench | 43.5% | — |
Math GPT-5 leads
Claude 3.5 Haiku: 13.7 (#294), GPT-5: 54.6 (#36)
| Benchmark | Claude 3.5 Haiku | GPT-5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.3% | 91.4% |
| Omni-MATH | 22.4% | 64.7% |
| LMArena Math | 1244 | 1406 |
| MATH Level 5 | 46.4% | 98.1% |
| FrontierMath (Feb 2025 set) | 0.3% | 32.4% |
| FrontierMath (Tiers 1-3) | — | 55.4% |
| FrontierMath Tier 4 | — | 22% |
| ProofBench | — | 18% |
| LiveBench Math | 35.5% | — |
| FrontierMath Tier 4 (v1) | — | 12.5% |
Knowledge GPT-5 leads
Claude 3.5 Haiku: 18.7 (#276), GPT-5: 56.5 (#37)
| Benchmark | Claude 3.5 Haiku | GPT-5 |
|---|---|---|
| GPQA Diamond | 38.1% | 86.2% |
| MMLU-Pro | 60.5% | 86.3% |
| Confabulations | 36.7% | 10.3% |
| GPQA (HELM) | 36.3% | 79.2% |
| LMArena Expert | 1208 | 1420 |
| Humanity's Last Exam | — | 25.3% |
| SimpleQA Verified | — | 50.1% |
| Vectara Hallucination Rate | — | 14.7% |
| MMLU | 74.3% | — |
Multimodal GPT-5 leads
Claude 3.5 Haiku: 26.3 (#113), GPT-5: 46.1 (#12)
| Benchmark | Claude 3.5 Haiku | GPT-5 |
|---|---|---|
| LMArena Vision | 1092 | 1232 |
| GeoBench | 34% | 81% |
| VPCT | — | 66% |
Multilingual GPT-5 leads
Claude 3.5 Haiku: 40.2 (#213), GPT-5: 51.5 (#105)
| Benchmark | Claude 3.5 Haiku | GPT-5 |
|---|---|---|
| LMArena Non-English | 1238 | 1397 |
| LMArena Chinese | 1229 | 1423 |
| LMArena French | 1262 | 1409 |
| LMArena German | 1237 | 1415 |
| LMArena Japanese | 1175 | 1411 |
| LMArena Korean | 1172 | 1361 |
| LMArena Russian | 1253 | 1406 |
| LMArena Spanish | 1261 | 1398 |
Instruction Following GPT-5 leads
Claude 3.5 Haiku: 63.1 (#229), GPT-5: 74.0 (#107)
| Benchmark | Claude 3.5 Haiku | GPT-5 |
|---|---|---|
| IFEval | 79.2% | 87.5% |
| LMArena Instruction Following | 1241 | 1388 |
| LiveBench Instruction Following | 61.9% | — |
Long Context GPT-5 leads
Claude 3.5 Haiku: 38.3 (#195), GPT-5: 69.5 (#2)
| Benchmark | Claude 3.5 Haiku | GPT-5 |
|---|---|---|
| LMArena Longer Query | 1261 | 1399 |
| Fiction.LiveBench | — | 97.2% |
Writing & Preference GPT-5 leads
Claude 3.5 Haiku: 42.9 (#228), GPT-5: 63.5 (#63)
| Benchmark | Claude 3.5 Haiku | GPT-5 |
|---|---|---|
| LMArena Text | 1255 | 1406 |
| LMArena Creative Writing | 1233 | 1365 |
| Short-Story Creative Writing | 73.5% | 86% |
| EQ-Bench Creative Writing | 1146 | 1627 |
| WildBench | 76% | 85.7% |
| LMArena Multi-Turn | 1265 | 1426 |
| LiveBench Language | 35.4% | — |
Frequently asked questions
Is Claude 3.5 Haiku better than GPT-5?
GPT-5 is the stronger model overall, scoring 50.9 to 28.9 on the Noometry Index.
Is Claude 3.5 Haiku or GPT-5 better for coding?
GPT-5 scores higher on coding benchmarks: 50.7 versus 32.6 in the Noometry coding category.
How many benchmarks do Claude 3.5 Haiku and GPT-5 share?
38 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and GPT-5 has 69.