Anthropic released Claude Opus 4.8, the latest update to its flagship Large Language Model (LLM), on May 28, 2026. <cite index="11-28,11-29">The model is available across Claude products, the Claude API (claude-opus-4-8), Amazon Bedrock, Google Cloud Vertex AI, and Microsoft Foundry.</cite> The release arrives roughly six weeks after Opus 4.7 and continues Anthropic's accelerated point-release cadence following Opus 4.5 in November 2025.
Pricing and availability
<cite index="11-30,11-31">Claude Opus 4.8 pricing is $5 per million input tokens and $25 per million output tokens, unchanged from Opus 4.7. An optional fast mode runs at 2.5x speed for $10 / $50 per million tokens, three times cheaper than fast mode on previous Claude models.</cite> <cite index="11-32">Opus 4.8 supports a 1 million token input context window with up to 128K output tokens, matching Opus 4.7.</cite> <cite index="8-20">Pricing for regular use has not changed compared to Opus 4.7.</cite>
Benchmark results
Anthropic frames the release as incremental. <cite index="19-4,19-5,19-6">Anthropic released Claude Opus 4.8 on May 28, 2026. The framing is modest. Anthropic itself calls it "a modest but tangible improvement" over Opus 4.7, and the pricing is unchanged at $5 per million input tokens and $25 per million output tokens.</cite>
Reported scores include <cite index="11-33">88.6% vs 87.6% on SWE-bench Verified, 69.2% vs 64.3% on SWE-bench Pro, 82.2% vs 77.3% on MCP-Atlas, and 84.3% vs 79.3% on BrowseComp (single-agent)</cite> relative to Opus 4.7. <cite index="11-34">The +4.9 point jump on SWE-bench Pro is the real coding signal: SWE-bench Verified is approaching saturation, so the harder, less-saturated set is where headroom remains.</cite> Anthropic also reports a leading <cite index="11-15">74.6% on Terminal-Bench 2.1, 93.6% on GPQA Diamond, and a leading 1890 Elo on GDPval-AA</cite>, though Terminal-Bench is not directly comparable to prior scores due to a version change.
Honesty and alignment
<cite index="8-3,8-4,8-5,8-6">Anthropic claims the model is a "more effective collaborator" with improvements in agentic coding, multidisciplinary reasoning, agentic computer use, knowledge work, and agentic financial analysis. Testers have found Opus 4.8 to be "more reliable and sharper in its judgement" when doing agentic tasks, and the model also made gains in honesty. 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.</cite>
Product changes
Anthropic shipped several companion features alongside the model. <cite index="8-11,8-12,8-13,8-14">Dynamic workflows (research preview) - Claude can complete bigger tasks in Claude Code. It is able to plan work and run hundreds of parallel subagents in a single session. It is able to complete codebase-scale migrations across hundreds of thousands of lines of code. The feature is available for Claude Code for Enterprise, Team, and Max plans.</cite> <cite index="8-15,8-16,8-17">Effort control - In Claude.ai and Cowork, users can choose how much effort Claude puts into a response. With a lower setting, Claude will respond faster and use up rate limits more slowly. Opus 4.8 defaults to high effort, which Anthropic says is the best balance of quality and user experience.</cite> The Messages API also now accepts mid-task system entries, allowing developers to update Claude's instructions during a session.
Roadmap
<cite index="8-21,8-22">Anthropic is working on models that have the same capabilities as Opus 4.8 at a lower cost, and a new class of model that's even more intelligent than Opus. Anthropic says it has been developing safeguards for the Claude Mythos model it is testing with a small number of organizations, and it expects to be able to bring Mythos-class models to all customers "in the coming weeks."</cite>