← All benchmarks
Claude 4.5 Haiku (Reasoning)
Anthropic · Released 2025-10-15 · Independent benchmark scores from Artificial Analysis
Intelligence Index
29.6
Output speed
149.2
tokens/sec
Blended price
$2.00
per 1M tokens
Time to first token
15.24s
Model metadata
| Model ID | a6340098-d7ae-462d-b372-0a0a67fc44b4 |
|---|---|
| Slug | claude-4-5-haiku-reasoning |
| Release date | 2025-10-15 |
| Provider | Anthropic |
| Provider slug | anthropic |
| Output speed | 149.2 tok/s |
| Time to first token | 15.24s |
| Time to first answer | 15.24s |
| Blended price ($/1M) | $2.00 |
| Input price ($/1M) | $1.00 |
| Output price ($/1M) | $5.00 |
All evaluation scores (17)
| aime | — |
|---|---|
| aime 25 | 83.7% |
| artificial analysis coding index | 43.90 |
| artificial analysis intelligence index | 29.60 |
| artificial analysis math index | 83.70 |
| gpqa | 67.2% |
| hle | 9.7% |
| ifbench | 54.3% |
| lcr | 70.3% |
| livecodebench | 61.5% |
| math 500 | — |
| mmlu pro | 76.0% |
| scicode | 43.3% |
| tau banking | 9.1% |
| tau2 | 54.7% |
| terminalbench hard | 27.3% |
| terminalbench v2 1 | 44.2% |
Source: artificialanalysis.ai/models/claude-4-5-haiku-reasoning
Benchmark data from Artificial Analysis.