Model comparison
Claude 3 Haiku vs Llama 2-7B
Llama 2-7B is the stronger model overall, scoring 29.1 to 25.9 on the Noometry Index.
Last verified . 19 shared benchmarks.
MetaSummary
- They share 19 benchmarks with published results for both. Claude 3 Haiku scores higher in 5 categories and Llama 2-7B in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Llama 2-7B leads 30.7 to 9.8.
- Llama 2-7B has downloadable open weights; the other is API-only.
Side by side
| Claude 3 Haiku | Llama 2-7B | |
|---|---|---|
| Provider | Anthropic | Meta |
| Noometry Index | 25.9 | 29.1 |
| Released | 2024-03-07 | 2023-07-18 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 37 | 29 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 2-7B leads
Claude 3 Haiku: 26.4 (#325), Llama 2-7B: 29.2 (#307)
| Benchmark | Claude 3 Haiku | Llama 2-7B |
|---|---|---|
| LMArena Coding | 1199 | 1002 |
| WeirdML | 9.8% | — |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |
Reasoning Too close to call
Claude 3 Haiku: 16.3 (#307), Llama 2-7B: 15.7 (#312)
| Benchmark | Claude 3 Haiku | Llama 2-7B |
|---|---|---|
| LMArena Hard Prompts | 1174 | 1009 |
| Epoch Capabilities Index | 118.35 | 99.06 |
| WinoGrande | 74.2% | 69.2% |
| Kagi LLM Benchmark | 34.2% | — |
| Chess Puzzles | — | 0% |
| DTBench | 50.1% | — |
| LMCA | 8.8% | — |
| BIG-Bench Hard | — | 39.2% |
| ForecastBench | 53.2 | — |
| HellaSwag | — | 77.2% |
| LAMBADA | — | 73.3% |
| PIQA | — | 78.8% |
Math Llama 2-7B leads
Claude 3 Haiku: 9.8 (#319), Llama 2-7B: 30.7 (#233)
| Benchmark | Claude 3 Haiku | Llama 2-7B |
|---|---|---|
| LMArena Math | 1188 | 1042 |
| OTIS Mock AIME 2024-2025 | 1.8% | — |
| MATH Level 5 | 14.9% | — |
| GSM8K | — | 16.7% |
Knowledge Llama 2-7B leads
Claude 3 Haiku: 17.3 (#285), Llama 2-7B: 28.2 (#248)
| Benchmark | Claude 3 Haiku | Llama 2-7B |
|---|---|---|
| LMArena Expert | 1148 | 1036 |
| MMLU | 73.8% | 45.8% |
| GPQA Diamond | 36.3% | — |
| Confabulations | 34.2% | — |
| ARC (AI2) Challenge | — | 45.9% |
| BoolQ | — | 77.9% |
| OpenBookQA | — | 58.6% |
| TriviaQA | — | 73.7% |
Multimodal Not comparable
Claude 3 Haiku: 23.6 (#128), Llama 2-7B: —
| Benchmark | Claude 3 Haiku | Llama 2-7B |
|---|---|---|
| ScienceQA | 72% | 43.1% |
| LMArena Vision | 950 | — |
Multilingual Claude 3 Haiku leads
Claude 3 Haiku: 36.0 (#243), Llama 2-7B: 23.8 (#293)
| Benchmark | Claude 3 Haiku | Llama 2-7B |
|---|---|---|
| LMArena Non-English | 1178 | 973 |
| LMArena Chinese | 1155 | 973 |
| LMArena French | 1195 | 970 |
| LMArena German | 1174 | 978 |
| LMArena Russian | 1204 | 995 |
| LMArena Spanish | 1166 | 1007 |
| LMArena Japanese | 1102 | — |
| LMArena Korean | 1109 | — |
Instruction Following Claude 3 Haiku leads
Claude 3 Haiku: 61.3 (#247), Llama 2-7B: 50.8 (#298)
| Benchmark | Claude 3 Haiku | Llama 2-7B |
|---|---|---|
| LMArena Instruction Following | 1173 | 1006 |
Long Context Claude 3 Haiku leads
Claude 3 Haiku: 36.1 (#237), Llama 2-7B: 30.4 (#287)
| Benchmark | Claude 3 Haiku | Llama 2-7B |
|---|---|---|
| LMArena Longer Query | 1190 | 999 |
Writing & Preference Claude 3 Haiku leads
Claude 3 Haiku: 29.7 (#291), Llama 2-7B: 28.0 (#298)
| Benchmark | Claude 3 Haiku | Llama 2-7B |
|---|---|---|
| LMArena Text | 1195 | 1053 |
| LMArena Creative Writing | 1157 | 1033 |
| LMArena Multi-Turn | 1190 | 1029 |
| EQ-Bench Creative Writing | 717 | — |
Frequently asked questions
Is Claude 3 Haiku better than Llama 2-7B?
Llama 2-7B is the stronger model overall, scoring 29.1 to 25.9 on the Noometry Index.
Is Claude 3 Haiku or Llama 2-7B better for coding?
Llama 2-7B scores higher on coding benchmarks: 29.2 versus 26.4 in the Noometry coding category.
How many benchmarks do Claude 3 Haiku and Llama 2-7B share?
19 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Llama 2-7B has 29.