Opus 4.8

Extracted Attributes

• Developer

  Anthropic

• API Model ID

  claude-opus-4-8

• Release Date

  2026-05-28

• Context Window

  1,000,000 tokens

• Knowledge Cutoff

  2026-01-31

• Max Output Tokens

  128,000 tokens

• Regular Input Pricing

  $5 per million tokens

• Regular Output Pricing

  $25 per million tokens

Timeline

• Knowledge and training data cutoff for Claude Opus 4.8. (Source: Simon Willison's Weblog)

  2026-01-31

• Anthropic officially launches Claude Opus 4.8, making it available on AWS and the Claude API. (Source: Coursiv Blog)

  2026-05-28

Claude (language model)

Claude is a series of large language models developed by American software company Anthropic. Claude was released as an AI-based chatbot in March 2023. It is also used in AI-assisted software development. Claude is trained using "constitutional AI", a technique developed by Anthropic to improve ethical and legal compliance (AI alignment). Since Claude 3, each generation has typically been released in three sizes, from least to most capable: Haiku, Sonnet, and Opus. An additional model named Claude Mythos was released to a handful of companies in 2026. US federal agencies started phasing out the use of Claude after Anthropic refused to remove contractual prohibitions on the use of Claude for mass domestic surveillance and fully-autonomous weapons. Following the refusal, the Department of Defense (DoD) designated the company a "supply chain risk" and barred all U.S. private military contractors, suppliers, and partners from doing business with the firm. On March 26, 2026, a federal judge issued a temporary injunction against the DoD's designation.

Web Search Results

• Claude Opus 4.8 is now available on AWS | Artificial Intelligence

  # Claude Opus 4.8 is now available on AWS Today, we’re excited to announce the availability of Anthropic’s most advanced Opus model, Claude Opus 4.8, on Amazon Bedrock and the Claude Platform on AWS. Claude Opus 4.8 represents a meaningful step forward, delivering improvements across the workflows teams run in production, from agentic coding and deep knowledge work to multi-stage autonomous tasks that span hours of independent operation. With Claude Opus 4.8 on Amazon Bedrock you can build within your existing AWS environment, maintain enterprise security and regional data residency, and scale inference. Claude Opus 4.8 is also available through Claude Platform on AWS, giving you Anthropic’s native platform experience when regional data residency isn’t required. [...] In coding, Opus 4.8 is designed to navigate real codebases, plan before editing, and maintain context across long sessions. On multi-stage tasks, it can track dependencies and sustain coherence over extended runs. This same autonomy extends into agentic workflows, where it can handle complex dependency chains and multi-step tool use with reduced oversight, making it a strong fit for both customer-facing and internal agents. In professional work, Opus 4.8 synthesizes long, complex sources into structured deliverables such as briefs, analyses, and reports. ## Industry Use Cases [...] This post covers Opus 4.8’s improvements and practical guidance for AI engineers integrating the model into agentic systems and production inference workloads on Amazon Bedrock. See the documentation for Claude Platform on AWS. ## What makes Claude Opus 4.8 different Claude Opus 4.8 is designed to change what teams can hand off to Claude, with stronger performance across coding, agentic tasks, and professional work, and the consistency and autonomy intended for long-running production workflows. Opus 4.8 can hold a plan across stages, better track what it has done and what remains, and adjust course when something breaks rather than surfacing an error and stopping. This should lead to more predictable behavior at scale with lower output variance and fewer review cycles.

• Claude Opus 4.8 Release Date, Pricing & API | Coursiv Blog

  Release date: Claude Opus 4.8 launched on May 28, 2026. Best use cases: advanced software engineering, AI agents, complex document work, long-context analysis, and enterprise workflows. Context window: Anthropic describes Opus 4.8 as a hybrid reasoning model with a 1M context window. API model ID: developers can use `claude-opus-4-8` through the Claude API. Pricing: regular usage starts at $5 per million input tokens and $25 per million output tokens. Fast mode pricing:$10 per million input tokens and $50 per million output tokens. Claude Code: Opus 4.8 supports large coding tasks and dynamic workflows for eligible Claude Code plans. Cost optimization: prompt caching and batch processing can reduce costs for suitable workloads. [...] ## Claude Opus 4.8 vs Claude Opus 4.7# Opus 4.8 is a refinement of Opus 4.7 rather than a complete product reset. The main improvements are reliability, self-checking, agentic task performance, and collaboration quality. [...] > Quick clarification: despite the word “Opus,” this is not an audio codec or audio-production platform. Claude Opus 4.8 is a frontier AI model from Anthropic. ## Claude Opus 4.8 release notes: what Anthropic announced# Quick answer: Anthropic announced Claude Opus 4.8 on May 28, 2026 as an Opus 4.7 upgrade with stronger coding, agentic-task, and professional-work performance, plus effort control, Claude Code dynamic workflows, and cheaper fast mode.

• What's new in Claude Opus 4.8

  ###  Adaptive thinking With adaptive thinking enabled, Claude Opus 4.8 triggers reasoning only when it judges the turn needs it. On simple lookups and short agentic steps it responds directly; on complex multi-step problems it reasons before answering. This reduces wasted thinking tokens on bimodal workloads compared to Claude Opus 4.7 at the same effort level. As on Claude Opus 4.7, thinking is off unless you explicitly set `thinking: {type: "adaptive"}` in your request. ##  Behavior changes These are not API breaking changes but may require prompt updates. See Migrating to Claude Opus 4.8 for full guidance. [...] Python  ``` # Before (Opus 4.6 or earlier) thinking = {"type": "enabled", "budget_tokens": 32000} # After (Opus 4.7 and later) thinking = {"type": "adaptive"} output_config = {"effort": "high"} ``` ##  Capability improvements ###  Improvement areas Compared with Claude Opus 4.7, Claude Opus 4.8 targets behavioral improvements in: Long-horizon agentic coding, including better long-context handling, fewer compactions, and better compaction recovery. Reasoning effort calibration, with more reliable behavior at each effort level across a range of domains. Tool triggering, with fewer cases of skipping a tool call that the task required. ###  Adaptive thinking [...] # What's new in Claude Opus 4.8 - Claude API Docs and a lower 1,024-token minimum cacheable prompt length. ##  New model | Model | API model ID | Description | --- | Claude Opus 4.8 | claude-opus-4-8 | Anthropic's most capable Opus-tier model for complex reasoning, long-horizon agentic coding, and high-autonomy work | Claude Opus 4.8 supports the 1M token context window by default on the Claude API, Amazon Bedrock, and Vertex AI (200k on Microsoft Foundry), 128k max output tokens, adaptive thinking, and the same set of tools and platform features as Claude Opus 4.7. For complete pricing and specs, see the models overview. ##  New features ###  Mid-conversation system messages

• Claude Opus 4.8: “a modest but tangible improvement”

  Both the reliable knowledge cutoff and the training data cutoff are January 2026, the same as for 4.7. The context window is still 1,000,000 tokens, and the max output is 128,000 tokens. The What’s new in Claude Opus 4.8 document has some of the more interesting details. These caught my eye: > Mid-conversation system messages. Claude Opus 4.8 accepts `role: "system"` messages immediately after a user turn in the `messages` array (subject to placement rules). This lets you append updated instructions later in a long-running conversation without restating the full system prompt, which preserves prompt cache hits on the earlier turns and reduces input cost on agentic loops. [...] > One of the most prominent improvements in Opus 4.8 is its _honesty_. We train all our models to be honest---for instance, to avoid making claims that they can’t support. But a general problem with AI models is that they sometimes jump to conclusions, confidently claiming to have made progress in their work despite the evidence being thin. Early testers report that Opus 4.8 is more likely to flag uncertainties about its work and less likely to make unsupported claims. This is borne out in our evaluations, which show that Opus 4.8 is around four times less likely than its predecessor to allow flaws in code it has written to pass unremarked. That linked system card includes the following: [...] # Claude Opus 4.8: “a modest but tangible improvement” # Simon Willison’s Weblog Subscribe Sponsored by:Teleport — Prevent access bottlenecks. Unify identity. Teleport replaces fragmented identity and access tooling with a single identity layer that security teams trust, and engineers want to use. ## Claude Opus 4.8: “a modest but tangible improvement” 28th May 2026 Anthropic shipped Claude Opus 4.8 today. My favourite thing about it is this note in the release announcement: > Users will find Opus 4.8 to be a modest but tangible improvement on its predecessor. There’s still more to be done: we’re working on developing and releasing models that provide many of the same capabilities as Opus at a lower cost.

• Introducing Claude Opus 4.8 - Anthropic

  Video 1 ## A note on effort Opus 4.8 defaults to high effort, which we judge to be the best overall balance of quality and user experience. On coding tasks, this effort level spends a similar number of tokens as Opus 4.7’s default, but with better performance. Users can choose “extra” (“`xhigh`” in Claude Code) or “max,” and the model will spend more tokens to get better results; we recommend using “extra” for difficult tasks and long-running asynchronous workflows. We have increased rate limits in Claude Code to accommodate the higher token usage of higher effort levels; users can select whichever makes sense for their particular project. ## What’s next? [...] ## What’s next? Users will find Opus 4.8 to be a modest but tangible improvement on its predecessor. There’s still more to be done: we’re working on developing and releasing models that provide many of the same capabilities as Opus at a lower cost. Not only that, but we plan to release a new class of model with even higher intelligence than Opus. As part of Project Glasswing, a small number of organizations are currently using Claude Mythos Preview for cybersecurity work. Models of this capability level require stronger cyber safeguards before they can be generally released. We’re making swift progress on developing these safeguards and expect to be able to bring Mythos-class models to all our customers in the coming weeks. ## Availability [...] Dynamic workflows. This new feature, available in research preview, allows Claude to take on even bigger tasks in Claude Code. Claude can plan the work and then run hundreds of parallel subagents in a single session (and with Opus 4.8, the agents can run for even longer). It then verifies its outputs before reporting back to the user. For example, Claude Code with Opus 4.8 can now carry out codebase-scale migrations across hundreds of thousands of lines of code from kickoff to merge, with the existing test suite as its bar. You can read more about dynamic workflows—available in Claude Code for Enterprise, Team, and Max plans—in this post.