Model comparison
Claude 3.5 Haiku vs Llama 3-70B
Claude 3.5 Haiku and Llama 3-70B score almost the same on the Noometry Index (29.2 vs 28.8), so choose on price, context window or the category you care about most.
Last verified . 25 shared benchmarks.
MetaSummary
- They share 25 benchmarks with published results for both. Claude 3.5 Haiku scores higher in 5 categories and Llama 3-70B in 4 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Claude 3.5 Haiku leads 28.0 to 21.1.
- The biggest single-benchmark swing is MATH Level 5: 46.4% for Claude 3.5 Haiku and 22.6% for Llama 3-70B.
- Llama 3-70B has downloadable open weights; the other is API-only.
Side by side
| Claude 3.5 Haiku | Llama 3-70B | |
|---|---|---|
| Provider | Anthropic | Meta |
| Noometry Index | 29.2 | 28.8 |
| Released | 2024-10-22 | 2024-04-18 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 49 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 3-70B leads
Claude 3.5 Haiku: 32.9 (#265), Llama 3-70B: 35.8 (#218)
| Benchmark | Claude 3.5 Haiku | Llama 3-70B |
|---|---|---|
| BigCodeBench Instruct | 46.1% | 43.6% |
| LMArena Coding | 1286 | 1206 |
| BigCodeBench Complete | 59% | 54.5% |
| Aider Polyglot | 28% | — |
| SciCode | 27.4% | — |
| WeirdML | 30.7% | — |
| LiveBench Coding | 51.4% | — |
| CadEval | 32% | — |
| HumanEval+ | — | 72% |
| MBPP+ | — | 69% |
Agentic & Tool Use Claude 3.5 Haiku leads
Claude 3.5 Haiku: 28.0 (#95), Llama 3-70B: 21.1 (#139)
Reasoning Too close to call
Claude 3.5 Haiku: 17.7 (#290), Llama 3-70B: 18.0 (#288)
| Benchmark | Claude 3.5 Haiku | Llama 3-70B |
|---|---|---|
| LMArena Hard Prompts | 1251 | 1195 |
| DTBench | 56.7% | 54.2% |
| Epoch Capabilities Index | 127.15 | 122.93 |
| Kagi LLM Benchmark | — | 35.1% |
| CritPt | 0% | — |
| LiveBench Reasoning | 28.1% | — |
| LiveBench Data Analysis | 48.5% | — |
| ForecastBench | — | 57.1 |
| LiveBench | 43.5% | — |
| WinoGrande | — | 83.5% |
Math Claude 3.5 Haiku leads
Claude 3.5 Haiku: 14.7 (#300), Llama 3-70B: 12.8 (#305)
| Benchmark | Claude 3.5 Haiku | Llama 3-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.3% | 4.3% |
| LMArena Math | 1244 | 1218 |
| MATH Level 5 | 46.4% | 22.6% |
| Omni-MATH | 22.4% | — |
| LiveBench Math | 35.5% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Llama 3-70B leads
Claude 3.5 Haiku: 18.7 (#281), Llama 3-70B: 20.8 (#277)
| Benchmark | Claude 3.5 Haiku | Llama 3-70B |
|---|---|---|
| GPQA Diamond | 38.1% | 40.6% |
| LMArena Expert | 1208 | 1149 |
| MMLU | 74.3% | 79.3% |
| MMLU-Pro | 60.5% | — |
| Confabulations | 36.7% | — |
| GPQA (HELM) | 36.3% | — |
Multimodal Not comparable
Claude 3.5 Haiku: 26.8 (#117), Llama 3-70B: —
| Benchmark | Claude 3.5 Haiku | Llama 3-70B |
|---|---|---|
| LMArena Vision | 1092 | — |
| GeoBench | 34% | — |
Multilingual Claude 3.5 Haiku leads
Claude 3.5 Haiku: 40.0 (#218), Llama 3-70B: 33.6 (#251)
| Benchmark | Claude 3.5 Haiku | Llama 3-70B |
|---|---|---|
| LMArena Non-English | 1238 | 1142 |
| LMArena Chinese | 1229 | 1114 |
| LMArena French | 1264 | 1232 |
| LMArena German | 1237 | 1169 |
| LMArena Japanese | 1175 | 1017 |
| LMArena Korean | 1173 | 1017 |
| LMArena Russian | 1253 | 1159 |
| LMArena Spanish | 1261 | 1241 |
Instruction Following Too close to call
Claude 3.5 Haiku: 62.9 (#234), Llama 3-70B: 62.5 (#238)
| Benchmark | Claude 3.5 Haiku | Llama 3-70B |
|---|---|---|
| LMArena Instruction Following | 1241 | 1194 |
| LiveBench Instruction Following | 61.9% | — |
| IFEval | 79.2% | — |
Long Context Claude 3.5 Haiku leads
Claude 3.5 Haiku: 38.3 (#200), Llama 3-70B: 35.6 (#240)
| Benchmark | Claude 3.5 Haiku | Llama 3-70B |
|---|---|---|
| LMArena Longer Query | 1261 | 1174 |
Writing & Preference Too close to call
Claude 3.5 Haiku: 42.7 (#234), Llama 3-70B: 42.8 (#231)
| Benchmark | Claude 3.5 Haiku | Llama 3-70B |
|---|---|---|
| LMArena Text | 1255 | 1221 |
| LMArena Creative Writing | 1233 | 1210 |
| LMArena Multi-Turn | 1265 | 1223 |
| Short-Story Creative Writing | 73.5% | — |
| EQ-Bench Creative Writing | 1146 | — |
| WildBench | 76% | — |
| LiveBench Language | 35.4% | — |
Frequently asked questions
Is Claude 3.5 Haiku better than Llama 3-70B?
Claude 3.5 Haiku and Llama 3-70B score almost the same on the Noometry Index (29.2 vs 28.8), so choose on price, context window or the category you care about most.
Is Claude 3.5 Haiku or Llama 3-70B better for coding?
Llama 3-70B scores higher on coding benchmarks: 35.8 versus 32.9 in the Noometry coding category.
How many benchmarks do Claude 3.5 Haiku and Llama 3-70B share?
25 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and Llama 3-70B has 31.