Claude Haiku 4.5
Anthropic·Oct 2025reasoningproprietary
Claude Haiku 4.5 delivers what was frontier coding performance five months prior at a third of the cost — it scores 73.3% on SWE-bench Verified, ahead of many flagship models from early 2025. It is the standard pick for sub-agents in multi-agent systems, high-volume chat, and latency-sensitive products built on the Claude stack. It was also the first Haiku with extended thinking.
Benchmark results
Where it shines
- Near-frontier coding at budget pricing
- Low latency — ideal for sub-agents and parallel task fan-out
- Extended thinking support in the fast tier
Alternatives to Claude Haiku 4.5
Alibaba's trillion-parameter API flagship — frontier-adjacent quality with strong agentic tool use at mid-tier prices.
OpenAI's flagship reasoning model with a 1M-token context window, built for hard coding, science and long-horizon agentic work.
Google's November 2025 frontier breakout — 91.9% GPQA Diamond and 37.5% HLE made it the reasoning leader of its generation.
Google's 2025 workhorse flagship — first mainstream thinking model with a 1M context, still widely deployed.
Frequently asked questions
- How much does the Claude Haiku 4.5 API cost?
- Claude Haiku 4.5 costs $1 per million input tokens and $5 per million output tokens, with cached input at $0.1 per million. A workload of 10M input and 1.5M output tokens per month costs about $17.50.
- What is the context window of Claude Haiku 4.5?
- Claude Haiku 4.5 supports a context window of 200,000 tokens (200K), with up to 64K output tokens per response.
- Is Claude Haiku 4.5 open source?
- No — Claude Haiku 4.5 is a proprietary model available through Anthropic's API and partner platforms.
- What are the best alternatives to Claude Haiku 4.5?
- The closest alternatives by overall capability are Qwen3-Max, GPT-5.5, Gemini 3 Pro, Gemini 2.5 Pro. See the comparison pages for detailed head-to-head breakdowns.