Claude Haiku 4.5
Anthropic·
LLMssmallcloud
Overview
Ultra-fast, edge-capable model.
Capabilities and innovations
On-device potentialSub-10ms latencyCost efficiency
Benchmarks
| Benchmark | Score | Source |
|---|---|---|
| GPQA Diamond | 75.2% | unsourced |
| Humanity’s Last Exam (no tools) | 21.3% | unsourced |
| SWE-bench Verified | 51.4% | unsourced |
| MMLU | 87.2% | unsourced |
- SWE-bench Pro: not reported by the vendor
- MMLU-Pro: not reported by the vendor
- Terminal-Bench 2.x: benchmark did not exist at release
- MMMU-Pro: benchmark did not exist at release
Links
More from Anthropic
Data curated by AI Model Timeline. See the methodology for admission criteria, benchmark eras and source priority.