Anthropic launches Claude Haiku 5.5 at about 75% lower cost, its third 5.5 model in a month before planned IPO

Anthropic released Claude Haiku 5.5 on October 7, upgrading its smallest model from version 4.5 about a year after the previous Haiku release. The company called it "the cheapest, fastest, and most capable small model we've ever released," saying it costs around 75% less to run than Haiku 4.5 on average. The model is aimed at high-volume, cost-sensitive tasks.

List pricing is $0.10 per million input tokens and $0.50 per million output tokens for prompts up to 100,000 tokens, rising to $0.50 and $2.50 above that, compared with $1 and $5 for Haiku 4.5, according to Anthropic's pricing page. Anthropic pitches it for summaries, database queries and classification, as a subagent alongside Opus 5.5 and Sonnet 5.5 on coding work, and for speed-sensitive uses such as live customer support and browser use. On Anthropic's published benchmarks it scored 72.4% on OSWorld 2.1 (offline subset) and 39.2% on Terminal-Bench 4.0.

The release is the third Claude 5.5 model in about a month, following Opus and Sonnet upgrades, and was reported as part of a lineup expansion before Anthropic's planned IPO. Anthropic paired the launch with a 50% cut to Sonnet 5.5 cache-read prices, to $0.10 per million tokens from $0.20, which it says makes Sonnet 5.5 around 20% cheaper on most agentic work, and introduced monthly API credits for Max and Team subscribers. The model is also available through cloud partners including AMZN's AWS, GOOGL's Google Cloud and MSFT's Microsoft.

What to watch: whether the lower prices pull more high-volume API and agent workloads to Claude, and how price cuts on both its small and mid-tier models show up in the financial disclosures that come with the IPO process. IPO timing and valuation have not been confirmed.