Anthropic
Claude 3 Haiku
Claude 3 Haiku is Anthropic's fastest and most compact model for near-instant responsiveness. Quick and accurate targeted performance. See the launch announcement and benchmark results [here](https://www.anthropic.com/news/claude-3-haiku) #multimodal
- Input / 1M tokens
- $0.250
- Output / 1M tokens
- $1.25
- Context window
- 200K tokens
- Provider
- Anthropic
- Cached input / 1M
- $0.030
- Knowledge cutoff
- 2023-08-31
Performance
Median streaming throughput and first-token latency measured by Artificial Analysis.
- Output tokens / sec
- 131 t/s
- Time to first token
- 0.51s
Benchmarks
Intelligence, coding, and math indexes plus the underlying evaluation scores.
- Intelligence Index
- 12
- Coding Index
- 7
- Math Index
- —
- MMLU-Pro
- —
- GPQA
- 37.4%
- HLE
- 3.9%
- LiveCodeBench
- 15.4%
- SciCode
- 18.6%
- MATH-500
- 39.4%
- AIME
- 1.0%
Benchmarks via Artificial Analysis