Model comparison
Claude Haiku 4.5 vs Llama 2-70B
Claude Haiku 4.5 is the stronger model overall, scoring 39.3 to 24.1 on the Noometry Index.
Last verified . 23 shared benchmarks.
MetaSummary
- They share 23 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 8 categories and Llama 2-70B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Haiku 4.5 leads 43.0 to 7.5.
- The biggest single-benchmark swing is MATH Level 5: 96.4% for Claude Haiku 4.5 and 3.3% for Llama 2-70B.
- Llama 2-70B has downloadable open weights; the other is API-only.
Side by side
| Claude Haiku 4.5 | Llama 2-70B | |
|---|---|---|
| Provider | Anthropic | Meta |
| Noometry Index | 39.3 | 24.1 |
| Released | 2025-10-15 | 2023-07-18 |
| Weights | Proprietary | Open |
| Context window | 200K | — |
| Max output | 64K | — |
| Input $ / M tokens | $1 | — |
| Output $ / M tokens | $5 | — |
| Results tracked | 53 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.9 (#71), Llama 2-70B: 30.8 (#281)
| Benchmark | Claude Haiku 4.5 | Llama 2-70B |
|---|---|---|
| LMArena Coding | 1454 | 1079 |
| SWE-bench Verified (bash only) | 66.6% | — |
| LMArena WebDev | 1330 | — |
| SWE-bench Multilingual | 64.7% | — |
| SciCode | 43.3% | — |
| WeirdML | 45.4% | — |
| ALE-Bench | 653.48 | — |
Agentic & Tool Use Not comparable
Claude Haiku 4.5: 33.5 (#50), Llama 2-70B: —
| Benchmark | Claude Haiku 4.5 | Llama 2-70B |
|---|---|---|
| Terminal-Bench | 35.5% | — |
| Berkeley Function Calling Leaderboard | 68.7% | — |
| DeepResearch Bench | 45.5% | — |
| BALROG | 31.2% | — |
| ExploitBench | 13.7% | — |
| Vending-Bench 2 | 458.89 | — |
Reasoning Claude Haiku 4.5 leads
Claude Haiku 4.5: 15.1 (#310), Llama 2-70B: 13.7 (#324)
| Benchmark | Claude Haiku 4.5 | Llama 2-70B |
|---|---|---|
| LMArena Hard Prompts | 1420 | 1073 |
| DTBench | 73.6% | 41.6% |
| Epoch Capabilities Index | 142.42 | 113.59 |
| ForecastBench | 61.4 | 51.4 |
| ARC-AGI-2 | 4% | — |
| NYT Connections (extended) | 14.3% | — |
| ARC-AGI-1 | 47.7% | — |
| CritPt | 0% | — |
| Chess Puzzles | 8% | — |
| LMCA | 30.9% | — |
| BIG-Bench Hard | — | 64.9% |
| CommonsenseQA 2.0 | — | 50% |
| HellaSwag | — | 85.3% |
| LAMBADA | — | 78.9% |
| PIQA | — | 82.8% |
| WinoGrande | — | 80.2% |
Math Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.0 (#69), Llama 2-70B: 7.5 (#320)
| Benchmark | Claude Haiku 4.5 | Llama 2-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.7% | 0% |
| LMArena Math | 1395 | 1091 |
| MATH Level 5 | 96.4% | 3.3% |
| Omni-MATH | 56.1% | — |
| FrontierMath (Feb 2025 set) | 5.9% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
| GSM8K | — | 69.6% |
Knowledge Claude Haiku 4.5 leads
Claude Haiku 4.5: 37.6 (#146), Llama 2-70B: 7.4 (#305)
| Benchmark | Claude Haiku 4.5 | Llama 2-70B |
|---|---|---|
| GPQA Diamond | 71.2% | 26.3% |
| LMArena Expert | 1443 | 1039 |
| SimpleQA Verified | 13.2% | — |
| MMLU-Pro | 77.7% | — |
| Vectara Hallucination Rate | 9.8% | — |
| GPQA (HELM) | 60.5% | — |
| ARC (AI2) Challenge | — | 78.3% |
| BoolQ | — | 88.6% |
| MMLU | — | 69.9% |
| OpenBookQA | — | 60.2% |
| TriviaQA | — | 87.6% |
Multimodal Not comparable
Claude Haiku 4.5: 27.0 (#111), Llama 2-70B: —
| Benchmark | Claude Haiku 4.5 | Llama 2-70B |
|---|---|---|
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1420 | — |
Multilingual Claude Haiku 4.5 leads
Claude Haiku 4.5: 50.1 (#124), Llama 2-70B: 27.9 (#269)
| Benchmark | Claude Haiku 4.5 | Llama 2-70B |
|---|---|---|
| LMArena Non-English | 1376 | 1045 |
| LMArena Chinese | 1415 | 994 |
| LMArena French | 1408 | 1089 |
| LMArena German | 1377 | 1041 |
| LMArena Japanese | 1339 | 927 |
| LMArena Korean | 1346 | 964 |
| LMArena Russian | 1380 | 1083 |
| LMArena Spanish | 1420 | 1142 |
Instruction Following Claude Haiku 4.5 leads
Claude Haiku 4.5: 71.6 (#144), Llama 2-70B: 55.1 (#273)
| Benchmark | Claude Haiku 4.5 | Llama 2-70B |
|---|---|---|
| LMArena Instruction Following | 1414 | 1071 |
| IFEval | 80.1% | — |
Long Context Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.6 (#86), Llama 2-70B: 32.2 (#265)
| Benchmark | Claude Haiku 4.5 | Llama 2-70B |
|---|---|---|
| LMArena Longer Query | 1427 | 1062 |
Writing & Preference Claude Haiku 4.5 leads
Claude Haiku 4.5: 58.1 (#119), Llama 2-70B: 32.5 (#274)
| Benchmark | Claude Haiku 4.5 | Llama 2-70B |
|---|---|---|
| LMArena Text | 1396 | 1115 |
| LMArena Creative Writing | 1371 | 1075 |
| LMArena Multi-Turn | 1409 | 1088 |
| WildBench | 83.9% | — |
| EQ-Bench 4 | 1064 | — |
Frequently asked questions
Is Claude Haiku 4.5 better than Llama 2-70B?
Claude Haiku 4.5 is the stronger model overall, scoring 39.3 to 24.1 on the Noometry Index.
Is Claude Haiku 4.5 or Llama 2-70B better for coding?
Claude Haiku 4.5 scores higher on coding benchmarks: 43.9 versus 30.8 in the Noometry coding category.
How many benchmarks do Claude Haiku 4.5 and Llama 2-70B share?
23 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and Llama 2-70B has 35.