Model comparison
Claude 3.5 Haiku vs GPT-5.2
GPT-5.2 is the stronger model overall, scoring 54.1 to 28.9 on the Noometry Index.
Last verified . 25 shared benchmarks.
Summary
- They share 25 benchmarks with published results for both. Claude 3.5 Haiku scores higher in 0 categories and GPT-5.2 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.2 leads 59.8 to 13.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.3% for Claude 3.5 Haiku and 96.1% for GPT-5.2.
Side by side
| Claude 3.5 Haiku | GPT-5.2 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 28.9 | 54.1 |
| Released | 2024-10-22 | 2025-12-11 |
| Weights | Proprietary | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $1.75 |
| Output $ / M tokens | — | $14 |
| Results tracked | 49 | 67 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.2 leads
Claude 3.5 Haiku: 32.6 (#258), GPT-5.2: 52.2 (#32)
| Benchmark | Claude 3.5 Haiku | GPT-5.2 |
|---|---|---|
| WeirdML | 30.7% | 72.2% |
| LMArena Coding | 1287 | 1447 |
| SWE-bench Verified | — | 73.8% |
| SWE-bench Verified (bash only) | — | 72.8% |
| Aider Polyglot | 28% | — |
| LMArena WebDev | — | 1415 |
| SWE-bench Multilingual | — | 66.7% |
| SciCode | 27.4% | — |
| GSO | — | 27.4% |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | 51.4% | — |
| BigCodeBench Complete | 59% | — |
| CadEval | 32% | — |
| ALE-Bench | — | 1,294 |
| AlgoTune | — | 2.05 |
Agentic & Tool Use GPT-5.2 leads
Claude 3.5 Haiku: 27.7 (#90), GPT-5.2: 40.0 (#24)
| Benchmark | Claude 3.5 Haiku | GPT-5.2 |
|---|---|---|
| Terminal-Bench | — | 64.9% |
| Berkeley Function Calling Leaderboard | — | 55.9% |
| GDPval | — | 49.7% |
| Remote Labor Index | — | 2.5% |
| τ²-bench Airline | — | 83% |
| τ²-bench Banking | — | 32.2% |
| τ²-bench Retail | — | 81.6% |
| τ²-bench Telecom | — | 89.7% |
| DeepResearch Bench | — | 41.1% |
| BALROG | 19.3% | — |
| LMArena Search | — | 1207 |
| METR Time Horizons | — | 75.3% |
| Vending-Bench 2 | — | 3,591 |
Reasoning GPT-5.2 leads
Claude 3.5 Haiku: 17.1 (#286), GPT-5.2: 50.1 (#33)
| Benchmark | Claude 3.5 Haiku | GPT-5.2 |
|---|---|---|
| LMArena Hard Prompts | 1251 | 1445 |
| DTBench | 56.7% | 90.9% |
| Epoch Capabilities Index | 127.15 | 153.45 |
| ARC-AGI-2 | — | 52.9% |
| SimpleBench | — | 45.8% |
| Kagi LLM Benchmark | — | 73.3% |
| NYT Connections (extended) | — | 83.6% |
| ARC-AGI-1 | — | 86.2% |
| CritPt | 0% | — |
| Chess Puzzles | — | 49% |
| EnigmaEval | — | 10.4% |
| EBR-Bench | — | 23% |
| LiveBench Reasoning | 28.1% | — |
| Mystery Game Puzzles | — | 23% |
| LiveBench Data Analysis | 48.5% | — |
| LMCA | — | 43.9% |
| ForecastBench | — | 60.1 |
| LiveBench | 43.5% | — |
Math GPT-5.2 leads
Claude 3.5 Haiku: 13.7 (#294), GPT-5.2: 59.8 (#32)
| Benchmark | Claude 3.5 Haiku | GPT-5.2 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.3% | 96.1% |
| LMArena Math | 1244 | 1439 |
| FrontierMath (Feb 2025 set) | 0.3% | 40.7% |
| FrontierMath (Tiers 1-3) | — | 67.4% |
| FrontierMath Tier 4 | — | 31.7% |
| MathArena Final-Answer Competitions | — | 72% |
| ProofBench | — | 15% |
| Omni-MATH | 22.4% | — |
| LiveBench Math | 35.5% | — |
| MATH Level 5 | 46.4% | — |
| FrontierMath Tier 4 (v1) | — | 18.8% |
Knowledge GPT-5.2 leads
Claude 3.5 Haiku: 18.7 (#276), GPT-5.2: 59.2 (#29)
| Benchmark | Claude 3.5 Haiku | GPT-5.2 |
|---|---|---|
| GPQA Diamond | 38.1% | 91.4% |
| LMArena Expert | 1208 | 1445 |
| Humanity's Last Exam | — | 27.8% |
| SimpleQA Verified | — | 37.1% |
| MMLU-Pro | 60.5% | — |
| Confabulations | 36.7% | — |
| Vectara Hallucination Rate | — | 8.4% |
| GPQA (HELM) | 36.3% | — |
| MMLU | 74.3% | — |
Multimodal GPT-5.2 leads
Claude 3.5 Haiku: 26.3 (#113), GPT-5.2: 50.9 (#7)
| Benchmark | Claude 3.5 Haiku | GPT-5.2 |
|---|---|---|
| LMArena Vision | 1092 | 1268 |
| GeoBench | 34% | — |
| VPCT | — | 84% |
| Furniture Assembly | — | 38.3% |
| LMArena Document | — | 1405 |
Multilingual GPT-5.2 leads
Claude 3.5 Haiku: 40.2 (#213), GPT-5.2: 53.6 (#63)
| Benchmark | Claude 3.5 Haiku | GPT-5.2 |
|---|---|---|
| LMArena Non-English | 1238 | 1425 |
| LMArena Chinese | 1229 | 1460 |
| LMArena French | 1262 | 1456 |
| LMArena German | 1237 | 1448 |
| LMArena Japanese | 1175 | 1421 |
| LMArena Korean | 1172 | 1390 |
| LMArena Russian | 1253 | 1440 |
| LMArena Spanish | 1261 | 1433 |
Instruction Following GPT-5.2 leads
Claude 3.5 Haiku: 63.1 (#229), GPT-5.2: 74.9 (#83)
| Benchmark | Claude 3.5 Haiku | GPT-5.2 |
|---|---|---|
| LMArena Instruction Following | 1241 | 1417 |
| LiveBench Instruction Following | 61.9% | — |
| IFEval | 79.2% | — |
Long Context GPT-5.2 leads
Claude 3.5 Haiku: 38.3 (#195), GPT-5.2: 44.0 (#74)
| Benchmark | Claude 3.5 Haiku | GPT-5.2 |
|---|---|---|
| LMArena Longer Query | 1261 | 1428 |
| CL-bench | — | 18.2% |
Writing & Preference GPT-5.2 leads
Claude 3.5 Haiku: 42.9 (#228), GPT-5.2: 67.0 (#30)
| Benchmark | Claude 3.5 Haiku | GPT-5.2 |
|---|---|---|
| LMArena Text | 1255 | 1439 |
| LMArena Creative Writing | 1233 | 1402 |
| EQ-Bench Creative Writing | 1146 | 1703 |
| LMArena Multi-Turn | 1265 | 1458 |
| Short-Story Creative Writing | 73.5% | — |
| WildBench | 76% | — |
| LiveBench Language | 35.4% | — |
Frequently asked questions
Is Claude 3.5 Haiku better than GPT-5.2?
GPT-5.2 is the stronger model overall, scoring 54.1 to 28.9 on the Noometry Index.
Is Claude 3.5 Haiku or GPT-5.2 better for coding?
GPT-5.2 scores higher on coding benchmarks: 52.2 versus 32.6 in the Noometry coding category.
How many benchmarks do Claude 3.5 Haiku and GPT-5.2 share?
25 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and GPT-5.2 has 67.