Anthropic Releases Claude Haiku 5.5 Small Model and Halves Sonnet 5.5 Cache Read Prices
AI Executive Summary
Anthropic PBC released Claude Haiku 5.5, a small model targeted at repetitive tasks, high-volume summaries, and subagent workloads, pricing it at 10 cents per million input token and 50 cents per million output token for prompt up to 100,000 token.
The company also halved cache read prices for Claude Sonnet 5.5 from 20 cents to 10 cents per million token.
Early testers like Asana Inc.
noted task completion latency over 30% lower and inference up to 2.5 times faster when using Haiku 5.5 for AI Teammates.
Why It Matters
Strategic TakeawayAggressive pricing compression in small frontier language models, paired with reduced context caching costs, accelerates the economic viability of multi-agent orchestration architectures by making subagent handoffs computationally inexpensive.
Multi-Vector Implications
- TECHNICALHaiku 5.5 introduces an adjustable effort setting and improved tokenizer efficiency, matching OpenAI GPT-6 Luna pricing while outperforming it on OSWorld 2.1 and Terminal-Bench 4.0 benchmarks.
- MARKETPricing parity at 10 cents per million input token intensifies the cost-per-token price war among foundation model providers, directly pressuring low-end inference margins.
- GOVERNANCEExpanded cybersecurity safeguards allow permissive defensive workflows while restricting penetration testing tools, managed via Anthropic's Cyber Verification Program.
Strategic Outlook
12-18M HorizonOver the next 12-18 months, foundation model providers will integrate adjustable effort settings and ultra-low-cost tiered pricing as standard feature for small models, shifting enterprise adoption decisively toward high-frequency subagent delegation frameworks.
Referenced Coverage & Sources
Read the full coverage below for original reporting, technical benchmarks, and complete primary source details.
Google Releases Gemini 4 Argon, Called Its Most Powerful Model yet
Google has released its latest Gemini model, marketing it as a workhorse for coding and cybersecurity work.
Google Announces Gemini 4 Argon AI Model, but You Can't Use It yet
So much for Gemini 3.5 Pro.
LWiAI Podcast #258 - Opus 5.5, Sol and Luna, Muse, DeepSeek-V4.1-Flash, Xi
(Belated post :/ ) Anthropic releases Opus 5.5 with lower prices and Fable-level performance, OpenAI launches GPT-6 Sol and Luna, boasting lower cost and fewer.
Google Announces Gemini 4 and Says It's so Capable That Only 'trusted Cyber Defenders' Can Have It Right Now
Google today revealed its next AI frontier model, which it's calling Gemini 4 Argon.
Claude
Claude is a family of state-of-the-art Large Language Models developed by Anthropic. Highly regarded for its reasoning, coding capabilities, and context window size, Claude models are trained using a methodology called Constitutional AI.
Anthropic
Anthropic is an AI safety and research company, creators of the Claude LLM family, founded by former OpenAI researchers to build steerable, reliable, and constitutional AI systems.
CAC
CAC (Customer Acquisition Cost) is the total cost required to acquire a new customer, including all sales and marketing expenses.
Explore technical glossaries, weekly market briefings, and editorial research articles related to this story:
Get top 5 high-signal AI news, venture funding rounds, and research papers auto-routed to dedicated channels every 3 hours.