Claude 4.5 Haiku (Reasoning)
Intelligence
37.1
#108/521
Coding
32.6
#113/427
Math
83.7
#52/265
Speed
126 tok/s
TTFT: 11.22s
Pricing
$1.00 / $5.00
per 1M tokens (in/out)
Claude 4.5 Haiku (Reasoning) is Anthropic’s latest AI model designed for advanced reasoning tasks. It processes at 125.501 tokens per second and is priced at $1 per million input tokens and $5 per million output tokens, targeting professional users.
When to Use Claude 4.5 Haiku (Reasoning)
✓ Best For
- Complex problem-solving tasks.
- Mathematical computations.
- Software development and coding assistance.
✗ Not Ideal For
- Basic conversational applications.
- High-speed real-time processing needs.
How Claude 4.5 Haiku (Reasoning) Compares
Intelligence Index · Higher is better
CohereAlibabaAnthropic
Benchmark Profile
Coding Index
MiniMaxOpenAIAnthropicGoogleZ AI
Output Speed · tok/s
MistralAlibabaAnthropicAmazonXiaomi
Math Index
AlibabaAnthropicOpenAIInclusionAI
Intelligence · Coding · Math
Intelligence
Coding
Math
All Benchmark Scores (13)
| Benchmark | Score | |
|---|---|---|
| Intelligence Index | 37.1 | |
| Coding Index | 32.6 | |
| Math Index | 83.7 | |
| MMLU-Pro | 76% | |
| GPQA | 672% | |
| LiveCodeBench | 615% | |
| HLE | 97% | |
| SciCode | 43.3% | |
| IFBench | 54.3% | |
| LCR | 70.3% | |
| TerminalBench Hard | 27.3% | |
| Tau2 | 54.7% | |
| AIME 2025 | 83.7% |
Data: Artificial Analysis · Updated: April 9, 2026
Frequently Asked Questions (15)
When was Claude 4.5 Haiku (Reasoning) released?
Claude 4.5 Haiku (Reasoning) was released on October 15, 2025.
Who created Claude 4.5 Haiku (Reasoning)?
Claude 4.5 Haiku (Reasoning) was created by Anthropic.
How intelligent is Claude 4.5 Haiku (Reasoning)?
Claude 4.5 Haiku (Reasoning) scores 37 on the Artificial Analysis Intelligence Index, placing it above average among other reasoning models in a similar price tier (median: 31).
How fast is Claude 4.5 Haiku (Reasoning)?
Claude 4.5 Haiku (Reasoning) generates output at 113.9 tokens per second (based on Anthropic's API), which is well above average compared to other reasoning models in a similar price tier (median: 67.8 t/s).
What is the latency of Claude 4.5 Haiku (Reasoning)?
Claude 4.5 Haiku (Reasoning) has a time to first token (TTFT) of 15.58s (based on Anthropic's API), which is at the higher end compared to other reasoning models in a similar price tier (median: 2.64s).
How much does Claude 4.5 Haiku (Reasoning) cost?
Claude 4.5 Haiku (Reasoning) costs $1.00 per 1M input tokens (very competitive, median: $1.35) and $5.00 per 1M output tokens (better than average, median: $8.40), based on Anthropic's API.
What is Claude 4.5 Haiku (Reasoning) API pricing?
Claude 4.5 Haiku (Reasoning) costs $1.00 per 1M input tokens and $5.00 per 1M output tokens (based on Anthropic's API). For a blended rate (3:1 input to output ratio), this is $2.00 per 1M tokens. Pricing may vary by provider.
How verbose is Claude 4.5 Haiku (Reasoning)?
When evaluated on the Intelligence Index, Claude 4.5 Haiku (Reasoning) generated 87M output tokens, which is at the higher end compared to other reasoning models in a similar price tier (median: 30M).
Is Claude 4.5 Haiku (Reasoning) a reasoning model?
Yes, Claude 4.5 Haiku (Reasoning) is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.
What input modalities does Claude 4.5 Haiku (Reasoning) support?
Claude 4.5 Haiku (Reasoning) supports text and image input.
What output modalities does Claude 4.5 Haiku (Reasoning) support?
Claude 4.5 Haiku (Reasoning) supports text output.
Can Claude 4.5 Haiku (Reasoning) process images?
Yes, Claude 4.5 Haiku (Reasoning) supports image input and can analyze, describe, and answer questions about images.
Is Claude 4.5 Haiku (Reasoning) multimodal?
Yes, Claude 4.5 Haiku (Reasoning) is multimodal. It can process text and image input and generate text output.
What is the context window of Claude 4.5 Haiku (Reasoning)?
Claude 4.5 Haiku (Reasoning) has a context window of 200k tokens. This determines how much text and conversation history the model can process in a single request.
Is Claude 4.5 Haiku (Reasoning) open source?
No, Claude 4.5 Haiku (Reasoning) is proprietary. The model weights are not publicly available.
