Model by Anthropic

Claude Haiku 4.5

Claude Haiku 4.5 is Anthropic's small, fast model from late 2025, still available but now legacy, so it suits only pipelines already built around it.

Checked 2026-11-05

Claude Haiku 4.5 is Anthropic's small, fast model from late 2025, still available but now legacy, so it suits only pipelines already built around it. Anthropic released it on 15 October 2025. It has a 200K token context window and 64K tokens of output, takes text and images in, and uses manual extended thinking rather than adaptive thinking. The list price is $1 per million input tokens and $5 per million output tokens, matching OpenRouter's listing as of 5 November 2026. Haiku 5.5 now starts from $0.10 and $0.50 and has a 1M token context, so the newer model is cheaper and larger. Anthropic's retirement line for Haiku 4.5 reads not sooner than 15 October 2026, a date that has now passed, so check the deprecations page before you depend on it. The reliable knowledge cutoff is February 2025. You can reach it through the Claude API, Bedrock, Google Cloud, Microsoft Foundry or OpenRouter. We have not run it ourselves. We would only keep it for stable, high-volume classification already tuned to it. Figures come from Anthropic's model page and OpenRouter's list.

Mentioned here

Questions

What is the context window of Claude Haiku 4.5?

200,000 tokens, with up to 64,000 tokens of output.

How much does Claude Haiku 4.5 cost?

$1 per million input tokens and $5 per million output tokens, checked 31 October 2026.

Can I run Claude Haiku 4.5 on my own hardware?

No. The weights are not published, so you use it through the maker's API or a hosting provider.

Who makes Claude Haiku 4.5?

Anthropic makes it, and released it on 15 October 2025.

Reviews

Loading reviews…

Write a review

Your rating

Up to 1000 characters. Stars alone are enough.

Where you can run it

Compare hosts

  • OpenRouter

    OpenRouter routes more than 600 models through one API key.

    • Model gateway (router)
    • OpenAI compatible

Similar

  • Claude Opus 4.7

    Claude Opus 4.7 is Anthropic's April 2026 top-tier model for long-running agents, and it is now a legacy model: Anthropic points users to Opus 5.5.

    • Closed

    Anthropic, 1M context, $5 in / 1M

  • Claude Sonnet 4.6

    Claude Sonnet 4.6 is Anthropic's mid-priced model for everyday coding and agent work, and it is now a legacy model, so new projects should look at Sonnet 5.5 first.

    • Closed

    Anthropic, 1M context, $3 in / 1M

  • Claude Sonnet 5

    Claude Sonnet 5 is Anthropic's mid-tier model for agentic coding at a low price, and it fits developers who want near-Opus results without Opus bills.

    • Closed

    Anthropic, 1M context, $2 in / 1M

  • Claude Opus 4.8

    Claude Opus 4.8 is the older Opus that many agent setups still run on, and it suits teams that have not yet moved to the 5 series.

    • Closed

    Anthropic, 1M context, $5 in / 1M

  • Claude Opus 5

    Claude Opus 5 is Anthropic's former flagship for long coding and agent work, and it suits teams that already pinned it and want steady behaviour.

    • Closed

    Anthropic, 1M context, $5 in / 1M