Anthropic cuts Claude Haiku prices 90 percent, reshaping small-model economics
Source published ·Modelwire updated
Original coverage: The Decoder ↗·How Modelwire adds context
The development
Anthropic's Claude Haiku 5.5 signals intensifying competition in the cost-per-capability tier. Performance gains are substantial, with OSWorld scores nearly quintupling, yet the real story is pricing: up to 90 percent token cost reductions force rivals to recalibrate their own margins. However, a tokenizer redesign that increases token consumption per task partially offsets savings, suggesting efficiency gains matter as much as raw price cuts. For builders choosing between models, this reshapes the economics of scaling inference workloads and forces recalibration of cost-benefit calculations across the small-model segment.
Modelwire’s AI-generated summary of coverage from The Decoder.
Modelwire analysis
Analyst takeOur AI-generated reading of the wider context and the next developments to watch.
The tokenizer change is the buried qualifier here: a redesign that inflates token counts per task means the advertised 90 percent price cut overstates real-world savings for existing workloads, and any builder migrating from prior Haiku versions needs to remodel their cost assumptions from scratch before committing.
This release lands in a genuinely difficult moment for Anthropic's revenue picture. The story from October 5th about Meta and Microsoft pulling back from Claude described a structural customer concentration problem, and aggressive small-model pricing looks less like confidence and more like a response to that pressure: defend the developer base when enterprise anchors are drifting. The startup credit program covered on October 6th fits the same pattern, targeting builders who might absorb inference volume that large enterprise clients are no longer providing. Meanwhile the Ramp AI Index piece from October 2nd documented that enterprise buyers are already paying less per unit of AI output, which means Anthropic is cutting prices into a market that was already compressing margins across the board.
Watch whether OpenAI or Google matches Haiku 5.5's price points on their own small models within the next six weeks. If neither moves, it suggests the tokenizer inflation offset is widely understood and the effective price cut is smaller than headlines indicate.
This interpretation is generated from the summary above and the archive coverage cited below. Our methodology · Report an error
Coverage behind this analysis
These archive entries ground the connection in our analysis. They are ordered by source publication date, with links to our coverage and the original sources.
·The Decoder
Microsoft and Meta slash Claude spending as Anthropic becomes rival
Anthropic faces a critical customer concentration problem as its two largest enterprise clients shift spending away from Claude. Microsoft reduced per-employee AI budgets tenfold within its cloud division, while Meta simultaneously halved its Claude Code deployment. The pullback reflects a broader competitive realignment: both companies are prioritizing internal AI capabilities over third-party models, transforming Anthropic…
MentionsAnthropic · Claude Haiku 5.5 · Claude Haiku · OSWorld
How this coverage is produced
Modelwire uses AI to generate summaries and context from source headlines, snippets, and selected archive coverage. Automated checks do not verify every claim, and items are not routinely reviewed by a person before publication. Zacaria Solis operates the site. Read the linked source for the full evidence and report errors through our corrections process.
Modelwire summarizes, we don’t republish. The Decoder originally reported this story as “Claude Haiku 5.5 arrives with massive price cuts proving the AI pricing arms race is far from over”. The full content lives on the-decoder.com. If you’re a publisher and want a different summarization policy for your work, see our takedown page.