# News: Anthropic launches Claude Haiku 5.5 with prices up to 90% below Haiku 4.5

> Anthropic released Claude Haiku 5.5 on 7 October 2026, pricing it 90% below Haiku 4.5 for requests up to 100,000 tokens and 50% below above that, which it puts at around 75% cheaper to run on average. It is the first Haiku-class model with an adjustable effort setting, and Anthropic positions it for high-volume work such as summaries, compaction and classification, as a subagent beside Opus 5.5 and Sonnet 5.5 on coding, and for speed-sensitive tasks like live support and browser use. Published benchmarks include Terminal-Bench 4.0 at 39.2% (Haiku 4.5 scored 0.0%) and OSWorld 2.1 offline subset at 72.4% (15.7%). Alongside the launch, Sonnet 5.5 cache-read prices were halved (about 20% cheaper on most agentic work), Max and Team subscribers get new monthly API credits ($100 at Max 5x, $200 at Max 20x, up to $500 pooled for teams), and the Python and TypeScript SDKs add beta computer-use and browser-use support.

Human page: https://testedactually.com/news/claude-haiku-5-5-launch/
Mirror: https://testedactually.com/news/claude-haiku-5-5-launch.md
Published: 2026-10-09

## What changed

Anthropic released Claude Haiku 5.5 on 7 October 2026, pricing it 90% below Haiku 4.5 for requests up to 100,000 tokens and 50% below above that, which it puts at around 75% cheaper to run on average. It is the first Haiku-class model with an adjustable effort setting, and Anthropic positions it for high-volume work such as summaries, compaction and classification, as a subagent beside Opus 5.5 and Sonnet 5.5 on coding, and for speed-sensitive tasks like live support and browser use. Published benchmarks include Terminal-Bench 4.0 at 39.2% (Haiku 4.5 scored 0.0%) and OSWorld 2.1 offline subset at 72.4% (15.7%). Alongside the launch, Sonnet 5.5 cache-read prices were halved (about 20% cheaper on most agentic work), Max and Team subscribers get new monthly API credits ($100 at Max 5x, $200 at Max 20x, up to $500 pooled for teams), and the Python and TypeScript SDKs add beta computer-use and browser-use support.

## What it means for buyers

Usage-billed coding assistants get a much cheaper high-volume workhorse: the model ships in Claude Code, and the halved Sonnet 5.5 cache-read price compounds the cut for agentic stacks, so buyers on Claude-backed tools in our ai-coding-assistants comparison should watch token costs fall through to credits and usage lines. The benchmark jump over Haiku 4.5 also removes most of the quality penalty for downshifting subagents to a small model, which changes the cost maths for agent-heavy workflows. New monthly credits make small API experiments effectively free on Max and Team plans.

## Sources

- [1] Anthropic: Introducing Claude Haiku 5.5 - https://www.anthropic.com/claude-haiku-5-5 (launch post: prices, benchmarks, credits, SDK changes, checked 2026-10-09)
- [2] Simon Willison: Claude Haiku 5.5 - https://simonwillison.net/2026/Oct/7/claude-haiku-5-5/ (independent confirmation of the launch and the API credit scheme, 7 October 2026)

TestedActually tracks pricing and programme changes as dated, sourced events: https://testedactually.com/news/.