Model comparison
Claude 2 vs Claude 3 Haiku
Claude 3 Haiku is the stronger model overall, scoring 25.7 to 24.7 on the Noometry Index.
Last verified . 7 shared benchmarks.
Summary
- They share 7 benchmarks with published results for both. Claude 2 scores higher in 1 category and Claude 3 Haiku in 2 categories; one gap is clear of the uncertainty.
- The widest gap is in reasoning, where Claude 2 leads 20.8 to 15.7.
Side by side
| Claude 2 | Claude 3 Haiku | |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Noometry Index | 24.7 | 25.7 |
| Released | 2023-07-11 | 2024-03-07 |
| Weights | Proprietary | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 8 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Claude 2: —, Claude 3 Haiku: 26.1 (#320)
| Benchmark | Claude 2 | Claude 3 Haiku |
|---|---|---|
| HumanEval+ | 61.6% | 68.9% |
| WeirdML | — | 9.8% |
| BigCodeBench Instruct | — | 39.4% |
| LMArena Coding | — | 1199 |
| BigCodeBench Complete | — | 50.1% |
| CadEval | — | 12% |
| MBPP+ | — | 68.8% |
Reasoning Claude 2 leads
Claude 2: 20.8 (#213), Claude 3 Haiku: 15.7 (#303)
| Benchmark | Claude 2 | Claude 3 Haiku |
|---|---|---|
| DTBench | 51.9% | 50.1% |
| Epoch Capabilities Index | 120.08 | 118.3 |
| Kagi LLM Benchmark | — | 34.2% |
| LMArena Hard Prompts | — | 1174 |
| LMCA | — | 8.8% |
| ForecastBench | — | 53.2 |
| WinoGrande | — | 74.2% |
Math Too close to call
Claude 2: 8.7 (#314), Claude 3 Haiku: 9.2 (#313)
| Benchmark | Claude 2 | Claude 3 Haiku |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.5% | 1.8% |
| MATH Level 5 | 11.7% | 14.9% |
| LMArena Math | — | 1188 |
Knowledge Too close to call
Claude 2: 17.0 (#282), Claude 3 Haiku: 17.3 (#280)
| Benchmark | Claude 2 | Claude 3 Haiku |
|---|---|---|
| GPQA Diamond | 34.7% | 36.3% |
| MMLU | 78.5% | 73.8% |
| Confabulations | — | 34.2% |
| LMArena Expert | — | 1148 |
| TriviaQA | 87.5% | — |
Multimodal Not comparable
Claude 2: —, Claude 3 Haiku: 23.1 (#123)
| Benchmark | Claude 2 | Claude 3 Haiku |
|---|---|---|
| LMArena Vision | — | 950 |
| ScienceQA | — | 72% |
Multilingual Not comparable
Claude 2: —, Claude 3 Haiku: 36.1 (#238)
| Benchmark | Claude 2 | Claude 3 Haiku |
|---|---|---|
| LMArena Non-English | — | 1178 |
| LMArena Chinese | — | 1155 |
| LMArena French | — | 1194 |
| LMArena German | — | 1174 |
| LMArena Japanese | — | 1102 |
| LMArena Korean | — | 1108 |
| LMArena Russian | — | 1204 |
| LMArena Spanish | — | 1165 |
Instruction Following Not comparable
Claude 2: —, Claude 3 Haiku: 61.5 (#242)
| Benchmark | Claude 2 | Claude 3 Haiku |
|---|---|---|
| LMArena Instruction Following | — | 1174 |
Long Context Not comparable
Claude 2: —, Claude 3 Haiku: 36.1 (#232)
| Benchmark | Claude 2 | Claude 3 Haiku |
|---|---|---|
| LMArena Longer Query | — | 1191 |
Writing & Preference Not comparable
Claude 2: —, Claude 3 Haiku: 29.8 (#286)
| Benchmark | Claude 2 | Claude 3 Haiku |
|---|---|---|
| LMArena Text | — | 1195 |
| LMArena Creative Writing | — | 1157 |
| EQ-Bench Creative Writing | — | 717 |
| LMArena Multi-Turn | — | 1190 |
Frequently asked questions
Is Claude 2 better than Claude 3 Haiku?
Claude 3 Haiku is the stronger model overall, scoring 25.7 to 24.7 on the Noometry Index.
How many benchmarks do Claude 2 and Claude 3 Haiku share?
7 benchmarks have published results for both models. Claude 2 has 8 scored results on Noometry and Claude 3 Haiku has 37.