--- title: Anthropic releases Claude Haiku 5.5, its newest small model description: Anthropic has released Claude Haiku 5.5, describing it as its cheapest, fastest, and most capable small model to date. The announcement came via the @claudeai... canonical: https://daily.dev/posts/anthropic-releases-claude-haiku-5-5-its-newest-small-model-wli6uk40j twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Anthropic releases Claude Haiku 5.5, its newest small model | daily.dev og:description: Anthropic has released Claude Haiku 5.5, describing it as its cheapest, fastest, and most capable small model to date. The announcement came via the @claudeai... og:url: https://daily.dev/posts/anthropic-releases-claude-haiku-5-5-its-newest-small-model-wli6uk40j og:image: https://api.daily.dev/og/posts/WLi6uK40J.png og:image:alt: Anthropic releases Claude Haiku 5.5, its newest small model og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Anthropic releases Claude Haiku 5.5, its newest small model **[Collections](https://daily.dev/sources/collections)** · 3 min read · 0 upvotes · 0 comments ## Summary Anthropic has released Claude Haiku 5.5, describing it as its cheapest, fastest, and most capable small model to date. The announcement came via the @claudeai and @AnthropicAI accounts and was amplified by other social media users, but no full pricing comparison or independent benchmarks were included, leaving the simultaneous cost, speed, and capability claims unverified for now. ## Content Anthropic has released Claude Haiku 5.5, the first update to its smallest model in nearly a year. The main change is price. Under 100,000 tokens, it costs $0.10 per million input tokens and $0.50 per million output tokens, down from $1 and $5 for Haiku 4.5. Anthropic says the average saving is around 75%. Some Anthropic staff describe it as 10x cheaper for prompts under 100K tokens. ## The pricing catch The two numbers differ for a reason. Above 100K tokens the price rises 5x. Haiku 5.5 also uses a less generous tokenizer, about 1.25x more tokens than Haiku 4.5 for the same prompt, so some of the headline saving gets eaten. OpenAI's GPT-6 Luna is priced the same below the threshold but is cheaper above it. @simonw posted pelicans and a set of notes on the pricing. Anthropic also halved the cache-read price for Sonnet 5.5 to $0.10 per million tokens. Max and Team subscribers get monthly API credits of $100, $200, or up to $500 pooled. They match the subscription cost and don't roll over. ## What it can do Haiku 5.5 adds an effort setting, a first for Haiku, which trades cost against accuracy. Gains over Haiku 4.5 are large: - OSWorld 2.1 (computer use): 72.4%, up from 15.7% - GDPval-AA v2.1 (knowledge work): 1,620, up from 735 - Terminal-Bench 4.0 (agentic coding): 39.2%, up from 0% - Chartography (reading charts without tools): 46.4%, up from 6.4% It beats GPT-6 Luna on most measures, while trailing Sonnet 5.5 and Z.ai's GLM-5.3-Flash on some. Anthropic still recommends Sonnet 5.5 and Opus 5.5 for complex agentic coding. Asana reported over 30% lower task-completion latency and up to 2.5x faster inference per agent turn than its current model. Cognition says Haiku 5.5 scores 58.4% on FrontierCode 1.1 in Devin, ahead of Sonnet 5 at about an eighth of the cost per task. ## Sub-agent work The pitch from Anthropic and early users is the same: let Opus 5.5 or Sonnet 5.5 do the heavy thinking and hand high-volume work to Haiku. Anthropic suggests summaries, compactions and database queries. @mikeyk put it as Opus doing the thinking and Haiku the volume. @addyosmani says his team likes it as a subagent next to Opus 5.5. @lydiahallie finds it well suited to tasks that are mostly reading, small edits and tool calls. @mattshumer_ called Opus or Sonnet with Haiku sub-agents "very likely the new alpha." Cognition also offers it as a sidekick in Fusion, with Opus 5.5 as lead. @jerryjliu0 tested it for document parsing on OpenDocRouter, launched the same day. It runs roughly $1.2 per 1,000 pages and does well on tables and reading order for that price, but struggles with charts, semantic formatting and bounding boxes. ## Availability Haiku 5.5 is on the Claude Platform, Claude Code, AWS, Google Cloud and Microsoft Azure as `claude-haiku-5-5`. The Python and TypeScript SDKs have beta support for computer and browser use. It is also generally available in GitHub Copilot, rolling out gradually to Pro, Pro+, Max, Business and Enterprise users in VS Code, Visual Studio, Copilot CLI, the cloud agent, JetBrains IDEs, Xcode, Eclipse and more. GitHub says early testing showed it matching Claude Sonnet 5 on many coding tasks with significantly fewer tokens and steps. It's billed at provider list pricing under usage-based billing. Admins can control access through the model policy, and new models are enabled by default unless disabled. ## Questions this post answers ### What is Claude Haiku 5.5 and how is it different from previous Haiku models? Claude Haiku 5.5 is Anthropic's newest small language model, announced as the cheapest, fastest, and most capable small model the company has released. Anthropic claims lower average cost than prior versions, though a full pricing breakdown was not shared alongside the announcement, and independent benchmarks confirming the speed and capability claims were not yet available. _Developers choosing a small, cost-efficient model for production can follow daily.dev as benchmarks for Claude Haiku 5.5 emerge._ ## Community take How the wider developer community reacted, aggregated from 5 discussions and 31 comments across x (as of 2026-10-07). **TL;DR:** Reactions focus on the new pricing and the sub-agent pattern of pairing cheap Haiku with Opus/Sonnet for orchestration, with genuine enthusiasm about cost savings tempered by close scrutiny of the tokenizer change and the 100K-token price cliff. **Sentiment:** 45% positive · 45% mixed · 10% skeptical **The case for** - Using Haiku as a cheap, fast sub-agent for file-reading and grunt work while a bigger model plans/orchestrates is seen as a smart architecture. - Some report large inference cost cuts (e.g. ~70%) from using small models for sub-agents. - The low per-token pricing puts Haiku in range for real-time/interactive use cases. **The pushback** - The less generous tokenizer effectively pulls the 100K price cliff in to around 80K old-tokenizer tokens, eating into the advertised savings. - Crossing the 100K token threshold roughly 5x's the cost, so people warn to total whole task runs (including retries) rather than trust headline pricing. - Unpinned model aliases/configs can silently keep hitting older, worse, more expensive model versions for weeks. **By community** - x (mixed): Discussion mixes genuine pricing/architecture analysis with lighthearted pelican-benchmark banter and sub-agent orchestration enthusiasm. **Open questions** - Does a document-routing system choose a cheap vs. expensive model per page or per whole document when only part of it is chart-heavy? - How well does Haiku perform as a sub-agent on long-running tasks rather than short bursts? **Highlights** > @simonw 100k uncached input + 1k output is $0.0105 at the listed rates. 100,001 + the same output is about $0.0525. for a task comparison i'd total the whole run, retries included, then divide by accepted outputs. the calls that cross that cutoff deserve their own row > — [AlekVectis on x](https://x.com/AlekVectis/status/2107941490295222602) > @simonw the 1.25x tokenizer also pulls the 100k price cliff in to about 80k old-tokenizer tokens > — [pwguler on x](https://x.com/pwguler/status/2107939311106830735) > @simonw Had the same gap bite from the other side - an unpinned CLI alias in my fleet config kept hitting the old Haiku tag for weeks after a cheaper, better model shipped. Nothing forces a re-check unless you go looking for it. > — [MaximTitarenko on x](https://x.com/MaximTitarenko/status/2107941574772686906) > @mattshumer_ small models for sub-agents cut my inference bill ~70% last quarter honestly the alpha is just fanning out without flinching at the cost > — [nkchandupatla on x](https://x.com/nkchandupatla/status/2107933189775950136) > @mattshumer_ the win is what comes back up. Haiku does the messy reading and hands the big model a short summary, so the expensive context stays clean. that matters more than the price > — [FloppyBenchAI on x](https://x.com/FloppyBenchAI/status/2107929554987741423) **Source threads** - [x](https://x.com/jerryjliu0/status/2107940101217309041) · 0 points · 2 comments - [x](https://x.com/jerryjliu0/status/2107922393016717725) · 0 points · 0 comments - [x](https://x.com/cognition/status/2107942832921075726) · 0 points · 0 comments - [x](https://x.com/simonw/status/2107938675405586713) · 0 points · 14 comments - [x](https://x.com/mattshumer_/status/2107928142967152903) · 0 points · 15 comments ## Similar posts on daily.dev - [Anthropic launches new version of scaled-down “Haiku” model](https://daily.dev/posts/anthropic-launches-new-version-of-scaled-down-haiku-model-y3gj91kwb) · TechCrunch · 0 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#anthropic](https://daily.dev/tags/anthropic) [View this post on daily.dev](https://daily.dev/posts/anthropic-releases-claude-haiku-5-5-its-newest-small-model-wli6uk40j) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Anthropic releases Claude Haiku 5.5, its newest small model","url":"https://daily.dev/posts/anthropic-releases-claude-haiku-5-5-its-newest-small-model-wli6uk40j","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/anthropic-releases-claude-haiku-5-5-its-newest-small-model-wli6uk40j"},"datePublished":"2026-10-07T18:02:35.610Z","dateModified":"2026-10-07T21:16:37.700Z","description":"Anthropic has released Claude Haiku 5.5, describing it as its cheapest, fastest, and most capable small model to date. The announcement came via the @claudeai...","isAccessibleForFree":true,"articleSection":"Collections","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Collections","logo":"https://media.daily.dev/image/upload/s--fk_6ycEi--/f_auto,q_auto/v1780996001/logos/collections?_a=BAMAMiWQ0","url":"https://daily.dev/sources/collections"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/anthropic-releases-claude-haiku-5-5-its-newest-small-model-wli6uk40j","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"llm,anthropic","timeRequired":"PT3M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Collections","item":"https://daily.dev/sources/collections"},{"@type":"ListItem","position":3,"name":"Anthropic releases Claude Haiku 5.5, its newest small model"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/anthropic-releases-claude-haiku-5-5-its-newest-small-model-wli6uk40j#faq","mainEntity":[{"@type":"Question","name":"What is Claude Haiku 5.5 and how is it different from previous Haiku models?","acceptedAnswer":{"@type":"Answer","text":"Claude Haiku 5.5 is Anthropic's newest small language model, announced as the cheapest, fastest, and most capable small model the company has released. Anthropic claims lower average cost than prior versions, though a full pricing breakdown was not shared alongside the announcement, and independent benchmarks confirming the speed and capability claims were not yet available. Developers choosing a small, cost-efficient model for production can follow daily.dev as benchmarks for Claude Haiku 5.5 emerge."}}]} ```