Loading page content
Claude Haiku 4.5Closed
Tool use · Reasoning
Evaluation overview
Overall rank
#102
68.7 points · 201 models
Coding rank
#79
69.3 points · 181 models
Data coverage
3 items
3 sources · updated through 2026-09-02
Domain performance
Coding#79 / 181
69.3
Research#89 / 195
79.5
Long context#47 / 95
76.5
Evaluation highlights
All evaluationsBasic information
- Context (input / output)
- 200K / 64K
- Release date
- 2025-10
- Input modalities
- Text · Image
- Output modalities
- Text
- Reliable knowledge cutoff
- 2025-02
- Training data cutoff
- 2025-07
Availability
AnthropicFull pricing
- Input
- $1 /1M
- Output
- $5 /1M