Model comparison
Claude 3 Haiku vs GPT-4o mini
Claude 3 Haiku and GPT-4o mini score almost the same on the Noometry Index (25.9 vs 25.5), so choose on price, context window or the category you care about most.
Last verified . 33 shared benchmarks.
Summary
- They share 33 benchmarks with published results for both. Claude 3 Haiku scores higher in 2 categories and GPT-4o mini in 7 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-4o mini leads 39.5 to 29.7.
- The biggest single-benchmark swing is MATH Level 5: 14.9% for Claude 3 Haiku and 52.6% for GPT-4o mini.
Side by side
| Claude 3 Haiku | GPT-4o mini | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 25.9 | 25.5 |
| Released | 2024-03-07 | 2024-07-18 |
| Weights | Proprietary | Proprietary |
| Context window | — | 128K |
| Max output | — | 16K |
| Input $ / M tokens | — | $0.15 |
| Output $ / M tokens | — | $0.60 |
| Results tracked | 37 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude 3 Haiku leads
Claude 3 Haiku: 26.4 (#325), GPT-4o mini: 22.0 (#335)
| Benchmark | Claude 3 Haiku | GPT-4o mini |
|---|---|---|
| WeirdML | 9.8% | 11.8% |
| BigCodeBench Instruct | 39.4% | 46.1% |
| LMArena Coding | 1199 | 1290 |
| BigCodeBench Complete | 50.1% | 57.4% |
| HumanEval+ | 68.9% | 83.5% |
| MBPP+ | 68.8% | 72.2% |
| Aider Polyglot | — | 3.6% |
| LiveBench Coding | — | 43.1% |
| CadEval | 12% | — |
Agentic & Tool Use Not comparable
Claude 3 Haiku: —, GPT-4o mini: 27.5 (#101)
| Benchmark | Claude 3 Haiku | GPT-4o mini |
|---|---|---|
| BALROG | — | 17.4% |
Reasoning Claude 3 Haiku leads
Claude 3 Haiku: 16.3 (#307), GPT-4o mini: 8.7 (#347)
| Benchmark | Claude 3 Haiku | GPT-4o mini |
|---|---|---|
| Kagi LLM Benchmark | 34.2% | 28.8% |
| LMArena Hard Prompts | 1174 | 1267 |
| DTBench | 50.1% | 54.4% |
| LMCA | 8.8% | 10.4% |
| Epoch Capabilities Index | 118.35 | 126.56 |
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 10.7% |
| Chess Puzzles | — | 0% |
| LiveBench Reasoning | — | 32.8% |
| Mystery Game Puzzles | — | 12% |
| LiveBench Data Analysis | — | 50% |
| ForecastBench | 53.2 | — |
| LiveBench | — | 41.3% |
| PIQA | — | 88.7% |
| WinoGrande | 74.2% | — |
Math Too close to call
Claude 3 Haiku: 9.8 (#319), GPT-4o mini: 10.4 (#314)
| Benchmark | Claude 3 Haiku | GPT-4o mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.8% | 6.9% |
| LMArena Math | 1188 | 1267 |
| MATH Level 5 | 14.9% | 52.6% |
| FrontierMath (Tiers 1-3) | — | 0.7% |
| Omni-MATH | — | 28% |
| LiveBench Math | — | 36.3% |
| GSM8K | — | 91.3% |
Knowledge Too close to call
Claude 3 Haiku: 17.3 (#285), GPT-4o mini: 17.7 (#284)
| Benchmark | Claude 3 Haiku | GPT-4o mini |
|---|---|---|
| GPQA Diamond | 36.3% | 37.7% |
| Confabulations | 34.2% | 37.2% |
| LMArena Expert | 1148 | 1235 |
| MMLU | 73.8% | 81.8% |
| SimpleQA Verified | — | 8.3% |
| MMLU-Pro | — | 60.3% |
| GPQA (HELM) | — | 36.8% |
| BoolQ | — | 88.7% |
Multimodal GPT-4o mini leads
Claude 3 Haiku: 23.6 (#128), GPT-4o mini: 25.9 (#122)
| Benchmark | Claude 3 Haiku | GPT-4o mini |
|---|---|---|
| LMArena Vision | 950 | 1066 |
| Video-MME | — | 64.8% |
| GeoBench | — | 64% |
| VPCT | — | 34% |
| ScienceQA | 72% | — |
Multilingual GPT-4o mini leads
Claude 3 Haiku: 36.0 (#243), GPT-4o mini: 42.0 (#199)
| Benchmark | Claude 3 Haiku | GPT-4o mini |
|---|---|---|
| LMArena Non-English | 1178 | 1266 |
| LMArena Chinese | 1155 | 1265 |
| LMArena French | 1195 | 1297 |
| LMArena German | 1174 | 1272 |
| LMArena Japanese | 1102 | 1216 |
| LMArena Korean | 1109 | 1195 |
| LMArena Russian | 1204 | 1275 |
| LMArena Spanish | 1166 | 1276 |
Instruction Following Too close to call
Claude 3 Haiku: 61.3 (#247), GPT-4o mini: 61.9 (#239)
| Benchmark | Claude 3 Haiku | GPT-4o mini |
|---|---|---|
| LMArena Instruction Following | 1173 | 1258 |
| LiveBench Instruction Following | — | 56.8% |
| IFEval | — | 78.2% |
Long Context GPT-4o mini leads
Claude 3 Haiku: 36.1 (#237), GPT-4o mini: 39.1 (#186)
| Benchmark | Claude 3 Haiku | GPT-4o mini |
|---|---|---|
| LMArena Longer Query | 1190 | 1289 |
Writing & Preference GPT-4o mini leads
Claude 3 Haiku: 29.7 (#291), GPT-4o mini: 39.5 (#248)
| Benchmark | Claude 3 Haiku | GPT-4o mini |
|---|---|---|
| LMArena Text | 1195 | 1286 |
| LMArena Creative Writing | 1157 | 1268 |
| EQ-Bench Creative Writing | 717 | 873 |
| LMArena Multi-Turn | 1190 | 1285 |
| Short-Story Creative Writing | — | 67.2% |
| WildBench | — | 79.1% |
| LiveBench Language | — | 28.6% |
Frequently asked questions
Is Claude 3 Haiku better than GPT-4o mini?
Claude 3 Haiku and GPT-4o mini score almost the same on the Noometry Index (25.9 vs 25.5), so choose on price, context window or the category you care about most.
Is Claude 3 Haiku or GPT-4o mini better for coding?
Claude 3 Haiku scores higher on coding benchmarks: 26.4 versus 22.0 in the Noometry coding category.
How many benchmarks do Claude 3 Haiku and GPT-4o mini share?
33 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and GPT-4o mini has 60.