Claude Haiku 4.5 (20251001) — benchmark results
October 1, 2025 Claude Haiku 4.5 snapshot row. Provider: Anthropic. Released 2025-10-01. Access: API.
Unified ELO 1585 ± 7, rank #515 of 1776 rated models, from 303 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| HELM AIR-Bench | 93.2 | Refusal Rate (%) | 100 |
| EuroEval Spanish NLU - Sentiment Headlines ES | 52.68 | Sentiment classification Score (%) | 97.3 |
| EuroEval Croatian NLU - MMS HR | 46.99 | Sentiment classification Score (%) | 97.2 |
| EuroEval Greek Summarization - Greek Wikipedia | 32.52 | Score (%) | 94.5 |
| EuroEval Spanish Common Sense Reasoning | 79.3 | Common Sense Reasoning Average Score (%) | 94.4 |
| EuroEval Portuguese NLU - SST-2 PT | 85.2 | Sentiment classification Score (%) | 93.9 |
| Gorilla API Bench (BFCL) | 68.7 | Overall Accuracy (%) | 93.9 |
| HAL GAIA Level 3 | 61.54 | Accuracy (%) | 93.8 |
| EuroEval French NLU - Allocine | 95.93 | Sentiment classification Score (%) | 92.3 |
| EuroEval Faroese NLU - FoQA | 76.16 | Reading comprehension Score (%) | 92.1 |
| EuroEval Dutch NLU - ScaLA NL | 63.67 | Linguistic acceptability Score (%) | 90.9 |
| EuroEval English Common Sense Reasoning | 84.31 | Common Sense Reasoning Average Score (%) | 90.9 |
Interactive version: theaggregate.ai/model?slug=claude-haiku-4-5-20251001 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.