Claude 3.5 Haiku
#174 of 274 scored.
AnthropicOct 22, 2024Closed
Capabilities Index
Range 120.7–129.5
127.2
Context
Tokens
—
Input
Per 1M tokens
—
Output
Per 1M tokens
—
Benchmarks
How it scores.
- GPQA Diamond17.5%
161st of 186
- Mock AIME4.2%
154th of 176
Everything else Epoch has for it
- Lech Mazur Writing73.5%
- MMLU65.7%
- MATH level 546.4%
- GeoBench34.0%
- CadEval32.0%
- WeirdML30.7%
- Aider polyglot28.0%
- DTBench27.8%
- Balrog19.3%
- FrontierMath-2025-02-28-Private0.6%
On the index
Around it.
Sources