Anthropic
Claude 4.5 Haiku (Reasoning)
Claude 4.5 Haiku (Reasoning) is an Anthropic Claude Haiku model, the faster and lighter Claude tier designed for responsive assistants, automation, and high-volume workflows. The benchmark view helps judge how much quality it retains relative to larger Claude variants while showing its speed and pricing tradeoffs.
Introducing Claude HaikuOperational Metrics
Output Speed103.8 tok/s
First Token12.26s
Blended Price$2.00/M
Model Metadata
Queryable facts extracted from the upstream model payload.
ReleaseOct 15, 2025
Context Windown/a
Modalitiesn/a
Strength: SciCode
Rank #53 across 472 models.
43.3%
Strength: Overall
Rank #93 across 500 models.
37.1
Strength: Math
Rank #52 across 269 models.
83.7
Watch Area: TTFT
Rank #262 across 293 models.
12.26s
Watch Area: Output $
Rank #228 across 325 models.
$5.00/M
Watch Area: Blended $
Rank #223 across 325 models.
$2.00/M
Strength Profile
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Benchmark Percentiles
Higher bars mean stronger relative placement.
All Benchmarks
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 37.1 | #93 |
| Artificial Analysis Coding Index | coding | 32.6 | #99 |
| Artificial Analysis Math Index | math | 83.7 | #52 |
| MMLU-Pro | reasoning | 76.0% | #166 |