16 October 2026

Anthropic cuts Haiku 5.5 prices and reports higher Terminal-Bench score

First reported

Deep Learning Weekly ran this on 16 October.

Anthropic, the company behind the Claude chatbot, lowered Haiku 5.5's per-token prices for requests under 100,000 tokens of context.
A token is roughly a word fragment, and a context window is how much text the model can hold at once.
On Terminal-Bench 4.0, a test of working through command-line tasks, Haiku 5.5 scored 39.2% at maximum effort.

Our read

Only Deep Learning Weekly, out of 30 newsletters, covered this story. It is the third Anthropic price-cut story in our archive after 2026-09-06 (Claude Haiku 5.5 with a 90% price cut) and 2026-10-05 (Anthropic and OpenAI cut AI model prices).

Earlier on this story

Oct 13Anthropic launches Claude-designed multi-agent workflows and a cheaper Haiku model
Oct 5Anthropic and OpenAI cut AI model prices, launch new versions
Sep 6Anthropic releases Claude Haiku 5.5 with a 90% price cut

Written by Newskeryx from which newsletters covered this story and from our own archive.

How it was covered

Deep Learning WeeklyEditorial team

Pricing tiers, and Terminal-Bench 4.0 scores against GPT-6 Luna