Claude Haiku 4.5 by Anthropic demonstrates fast output speed. View detailed benchmark data including scores across coding, math, reasoning, speed, and cost metrics.
Specifications last checked against the provider on . Benchmark scores carry their own source, linked beside each number.
General Benchmarks
Massive Multitask Language Understanding professional benchmark
HuggingFace LeaderboardArtificial Analysis composite intelligence score across 10 sub-benchmarks
Artificial Analysis (via OpenRouter)Human preference ELO from blind head-to-head votes
LMSYS / HuggingFaceCoding Benchmarks
Human preference ELO for building a working web page from a brief
Design Arena (via OpenRouter)Artificial Analysis composite coding score
Artificial Analysis (via OpenRouter)Human preference ELO for building things — websites, UI, games, charts, SVG
Design Arena (via OpenRouter)Math Benchmarks
Reasoning Benchmarks
Multi-turn service agent making tool calls under strict policy constraints
OpenRouter (measured)Artificial Analysis composite score for multi-step tool-using tasks
Artificial Analysis (via OpenRouter)Speed Benchmarks
Cost Benchmarks
Cost per 1M cached input tokens — the price that actually applies to a long agent conversation
OpenRouter APIContext Benchmarks
Available from 41 providers
The same model costs different amounts depending on who serves it. You can bring your own key for 6 of these — connect it here.
| Provider | Input / 1M | Output / 1M | Cached in | Context | Your key |
|---|---|---|---|---|---|
| QiHang | $0.14 | $0.71 | — | 200K | — |
| Xpersona | $0.6 | $3.7 | $0.06 | 200K | — |
| Poe | $0.85 | $4.3 | $0.085 | 192K | — |
| Cortecs | $0.996 | $4.982 | $0.099 | 200K | — |
| Abacus | $1 | $5 | — | 200K | — |
| Amazon Bedrock | $1 | $5 | $0.1 | 200K | Supported |
| Anthropic | $1 | $5 | $0.1 | 200K | Supported |
| Azure | $1 | $5 | $0.1 | 200K | Supported |
| Azure Cognitive Services | $1 | $5 | $0.1 | 200K | — |
| Cloudflare AI Gateway | $1 | $5 | $0.1 | 200K | — |
| CrossModel | $1 | $5 | $0.1 | 200K | — |
| DaoXE | $1 | $5 | $0.1 | 200K | — |
| Databricks | $1 | $5 | $0.1 | 200K | — |
| DevPass (LLM Gateway) | $1 | $5 | $0.1 | 200K | — |
| DigitalOcean | $1 | $5 | $1 | 200K | — |
| FreeModel | $1 | $5 | $0.1 | 200K | — |
| FrogBot | $1 | $5 | $0.1 | 200K | — |
| GitHub Copilot | $1 | $5 | $0.1 | 200K | — |
| Impossibl | $1 | $5 | $0.1 | 200K | — |
| Kilo Gateway | $1 | $5 | $0.1 | 200K | — |
| LLM Gateway | $1 | $5 | $0.1 | 200K | — |
| Merge Gateway | $1 | $5 | $0.1 | 200K | — |
| NanoGPT | $1 | $5 | $0.1 | 200K | — |
| NEAR AI Cloud | $1 | $5 | $0.1 | 200K | — |
| Neon | $1 | $5 | $0.1 | 200K | — |
| Ofox | $1 | $5 | $0.1 | 200K | — |
| OpenCode Zen | $1 | $5 | $0.1 | 200K | — |
| OpenRouter | $1 | $5 | $0.1 | 200K | Supported |
| Opper | $1 | $5 | $0.1 | 200K | — |
| OrcaRouter | $1 | $5 | $0.1 | 200K | — |
| Perplexity Agent | $1 | $5 | $0.1 | 200K | — |
| Pioneer | $1 | $5 | $0.1 | 200K | — |
| Requesty | $1 | $5 | $0.1 | 200K | — |
| Vercel AI Gateway | $1 | $5 | $0.1 | 200K | Supported |
| Vertex | $1 | $5 | $0.1 | 200K | Supported |
| Vertex (Anthropic) | $1 | $5 | $0.1 | 200K | — |
| ZenMux | $1 | $5 | $0.1 | 200K | — |
| UnoRouter | $1.2 | $6 | — | 200K | — |
| AnyAPI | — | — | — | 200K | — |
| Model Oracle AI | — | — | — | 200K | — |
| Snowflake Cortex | — | — | — | 200K | — |
Claude Haiku 4.5 — Benchmark Scores Overview
Scores normalized to percentage scale for visual comparison. ELO scores mapped to 0-100 range (1100-1500).
Compare Claude Haiku 4.5 With
Claude Haiku 4.5 — Frequently Asked Questions
How intelligent is Claude Haiku 4.5?
Claude Haiku 4.5 scores 1220 on the Chatbot Arena ELO rating, making it an entry-level AI model. This score is based on blind head-to-head human preference voting.
How much does Claude Haiku 4.5 cost?
Claude Haiku 4.5 costs $1.0 per 1M input tokens and $5.0 per 1M output tokens. This is mid-range pricing for its capability level.
How fast is Claude Haiku 4.5?
Claude Haiku 4.5 generates output at 180 tokens per second, which is very fast compared to other models. The time to first token is 120 ms.
How good is Claude Haiku 4.5 at coding?
Claude Haiku 4.5 achieves 30.0% on SWE-bench Verified, demonstrating moderate real-world software engineering capability. This benchmark tests the model's ability to resolve actual GitHub issues.
How good is Claude Haiku 4.5 at math and reasoning?
Claude Haiku 4.5 scores 70.0% on the MATH benchmark (competition-level mathematics). It also achieves 72.4% on GPQA Diamond, a graduate-level science reasoning benchmark.
What is the context window of Claude Haiku 4.5?
Claude Haiku 4.5 has a context window of 200K tokens. This determines how much text, conversation history, and code the model can process in a single request.
Who created Claude Haiku 4.5?
Claude Haiku 4.5 was created by Anthropic. It is classified as a budget model in our catalogue.
Is Claude Haiku 4.5 open source?
No, Claude Haiku 4.5 is a proprietary model. It is available through Anthropic's API and compatible providers.