This model is deprecated. We still benchmark its performance, but less often: every 18 hours instead of every 6.
Anthropic has launched a newer release, Claude Haiku 5.5. We suggest considering it instead.
Claude 4.5 Haiku: Intelligence, Performance & Price Analysis
Available providers
Model summary
Claude 4.5 Haiku (Reasoning) is among the least intelligent models, but well priced when comparing to other models of similar price. It's also notably quick per task and fairly concise. The model supports text and image input, outputs text, and has a 200k tokens context window with knowledge up to July 2025.
Claude 4.5 Haiku (Reasoning) scores 7 on the Artificial Analysis Intelligence Index, placing it at the lower end among comparable models (median: 41). When evaluating the Intelligence Index, it generated 40M tokens, which is fairly concise in comparison to the median of 78.6M.
Pricing for Claude 4.5 Haiku (Reasoning) is $1.00 per 1M input tokens (competitively priced, median: $2.00) and $5.00 per 1M output tokens (moderately priced, median: $10.00). On average, it costs $0.53 per task to evaluate Claude 4.5 Haiku (Reasoning) on the Intelligence Index.
At 3.7 minutes per Intelligence Index task, Claude 4.5 Haiku (Reasoning) is notably quick compared with the median of 13.3 minutes for comparable models (lower is better).
| Input modality | Supports: text and image |
|---|---|
| Output modality | Supports: text |
| Knowledge cutoff | Jul 1, 2025 |
| Context window | 200k ~300 A4 pages of size 12 Arial font |
Metrics are compared against models of the same class:
- Non-reasoning models → compared only with other non-reasoning models
- Reasoning models → compared across both reasoning and non-reasoning
- Open weights models → compared only with other open weights models of the same size class:
- Tiny: ≤4B parameters
- Small: 4B–40B parameters
- Medium: 40B–150B parameters
- Large: >150B parameters
- Proprietary models → compared across proprietary and open weights models of the same price range, using a blended 3:1 input/output price ratio:
- <$0.15 per 1M tokens
- $0.15–$1 per 1M tokens
- >$1 per 1M tokens
Intelligence
Artificial Analysis Intelligence Index
Artificial Analysis Intelligence Index by Open Weights / Proprietary
Intelligence Index vs. Cost per Task
Capability Indexes
Incorporates 5 evaluations: AA-Briefcase v1.1, AA-Omniscience v1.1, HLE-Diamond, AutomationBench-AA, GDP.pdf · Higher is better
Incorporates 4 evaluations: AutomationBench-AA, AA-Briefcase v1.1, AA-Omniscience v1.1, GDP.pdf · Higher is better
Incorporates 5 evaluations: AA-Omniscience v1.1, AA-Briefcase v1.1, HLE-Diamond, GDP.pdf, AutomationBench-AA · Higher is better
Incorporates 5 evaluations: AA-Omniscience v1.1, AA-Briefcase v1.1, MLCR-AA, AutomationBench-AA, HLE-Diamond · Higher is better
Incorporates 7 evaluations: AA-Omniscience v1.1, CritPt v2 (Hard Subset), Terminal-Bench-Science 0.1, HLE-Diamond, Terminal-Bench 4.0, FrontierCode-AA 2.0, AA-Briefcase v1.1 · Higher is better
Incorporates 4 evaluations: AA-Omniscience v1.1, AA-Briefcase v1.1, HLE-Diamond, GDP.pdf · Higher is better
Evaluations
Agentic knowledge work, Elo
Knowledge
1 - hallucination rate
Agentic SaaS workflows
Agentic scientific research workflows in a terminal
Agentic software engineering in open-source repositories
Reasoning & knowledge
Professional document reasoning, All-pass
Agentic coding & terminal use
Physics reasoning
Quantitative analysis on spreadsheets & documents
Agentic business operations
Agentic real-world work tasks, Elo
Kubernetes incident root-cause analysis
Medical long context reasoning
Visual reasoning
Cost
Cost per Intelligence Index Task
Cost to Run Artificial Analysis Intelligence Index
Pricing: Cache Hit, Input, and Output
Performance
Measured by time per task (minutes), output speed (tokens per second), time to first answer token and time to first token (seconds)
Time per Intelligence Index Task
Output Speed
Time to First Answer Token
Token Use
Output Tokens per Intelligence Index Task
Context Window
Context Window
Further details
Weights | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| Claude 4.5 Haiku (Reasoning) | 7 | - | 200k | $0.77 | 3.7 | 113 | Not available | ||
| Claude 4.5 Haiku (Non-reasoning) | - | - | 200k | $0.77 | - | 104 | Not available |
Frequently Asked Questions
Common questions about Claude 4.5 Haiku
Claude 4.5 Haiku was released on October 15, 2025.
Claude 4.5 Haiku was created by Anthropic.
Claude 4.5 Haiku has 2 effort levels: reasoning and non-reasoning. Claude 4.5 Haiku scores 7 (reasoning) on the Artificial Analysis Intelligence Index. Claude 4.5 Haiku (Reasoning) is at the lower end among other models in a similar price tier (median: 41).
Claude 4.5 Haiku takes 3.7 minutes (reasoning) per Intelligence Index task (based on Anthropic's API; lower is better). Claude 4.5 Haiku (Reasoning) is among the quickest compared to other models in a similar price tier (median: 13.3 minutes).
Claude 4.5 Haiku has a time to first answer token of 25.01s (reasoning) and 1.07s (non-reasoning) (based on Anthropic's API). Claude 4.5 Haiku (Reasoning)'s time to first answer token is somewhat higher than average compared to other models in a similar price tier (median: 12.24s).
Claude 4.5 Haiku costs $1.00 per 1M input tokens (very competitive, median: $2.00) and $5.00 per 1M output tokens (better than average, median: $10.00), based on Anthropic's API.
Claude 4.5 Haiku costs $1.00 per 1M input tokens and $5.00 per 1M output tokens (based on Anthropic's API). For a blended rate (7:2:1 cache hit/input/output ratio), this is $0.77 per 1M tokens. Pricing may vary by provider. Compare provider pricing
When evaluated on the Intelligence Index, Claude 4.5 Haiku generated 40M output tokens (reasoning). Claude 4.5 Haiku (Reasoning)'s output token count is better than average compared to other models in a similar price tier (median: 78.6M).
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. No, Claude 4.5 Haiku (Non-reasoning) is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.
Claude 4.5 Haiku supports text and image input.
Claude 4.5 Haiku supports text output.
Yes, Claude 4.5 Haiku supports image input and can analyze, describe, and answer questions about images.
Yes, Claude 4.5 Haiku is multimodal. It can process text and image input and generate text output.
Claude 4.5 Haiku has a context window of 200k tokens. This determines how much text and conversation history the model can process in a single request.
No, Claude 4.5 Haiku is proprietary. The model weights are not publicly available.
Claude 4.5 Haiku is a proprietary model and Anthropic has not disclosed the model size or parameter count.
Claude 4.5 Haiku achieves a score of 7 (reasoning) on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across knowledge work, coding, science, and general knowledge.
Claude 4.5 Haiku has a knowledge cutoff of July 2025. The model's training data includes information up to this date.
Yes, Claude 4.5 Haiku is available via API through 4 providers. Compare API providers
Claude 4.5 Haiku is available through 4 API providers. Compare providers