Claude Opus 4.8
Chat model · Anthropic ·
Claude Opus 4.8 is a chat model by Anthropic. In Endless it costs 405 credits per 1M input tokens and 2025 per 1M output tokens, on the Hobby plan and up.
Claude Opus 4.8 is Anthropic's May 2026 Opus model for complex coding and long agent tasks, more likely to flag doubts about its own work.
- Price
- 405 credits per 1M input tokens
- 2025 credits per 1M output tokens
- Availability
- Available on Hobby and above
- Context window
- 1M tokens
- Max output
- 128K tokens
- Inputs
- Text, images
- Outputs
- Text
About Claude Opus 4.8
Claude Opus 4.8 shipped in May 2026 as a modest but tangible upgrade over Claude Opus 4.7. Its main change is honesty about its own work: it flags uncertainties more often, makes fewer unsupported claims and is about four times less likely than Opus 4.7 to let a flaw in code it wrote pass without comment. It reads text and images with a 1M-token context window.
In agent work it asks the right questions, catches its own mistakes and pushes back when a plan isn't sound. That makes it a fit for agentic coding, large migrations split across many parallel subagents, legal tasks and analysis of documents and diagrams, where it also points out problems with the inputs and outputs of an analysis.
Claude Opus 5 and Claude Opus 5.5 have since followed it in the Opus line. Opus 4.8 still fits teams that validated their workflows on it and want the same behavior, and you can switch to a newer Opus in the same Endless workspace whenever you are ready.
Best for
- Code changes where you need the model to say what it is unsure of
- Checking an analysis for problems in its inputs and outputs
- Large migrations split into many parallel subtasks
- Legal and document-heavy agent work
Questions about Claude Opus 4.8
- What changed in Claude Opus 4.8 compared with Opus 4.7?
- It is a modest but tangible improvement: more reliable agentic work, fewer unsupported claims and far fewer unflagged flaws in the code it writes.
- Does Claude Opus 4.8 support a 1M-token context?
- Yes. It has a 1M-token context window and can write up to 128K tokens in a single response.
Related models
Claude Opus 5.5
by Anthropic
Long agentic coding and research work over a 1M-token context
324 / 1620 credits per 1M tokens
Claude Haiku 5.5
by Anthropic
Fast summaries, routing and subagent tasks at high volume
40.5 / 202.5 credits per 1M tokens
Claude Fable 5
by Anthropic
Long autonomous sessions that stay focused across millions of tokens
810 / 4050 credits per 1M tokens
Give your next idea a place to grow.
Start free, create a project, and bring your brand’s references together. Or book a demo to explore Endless with your team.