Skip to main content

Try Claude Opus 4.8 in the Workbench

Run this model interactively, tune parameters, and compare outputs.
Model ID: claude-opus-4-8 Claude Opus 4.8 is Anthropic’s most capable generally available model, designed for complex reasoning, long-horizon agentic coding, and high-autonomy professional work. It builds on Claude Opus 4.7 with gains across agentic coding, agentic terminal use, multidisciplinary reasoning, and computer use, while keeping a native 1M-token context window at standard pricing. Other noteworthy features of Claude Opus 4.8 include adaptive thinking with user-controllable effort levels (defaulting to high), computer use for multi-step workflows, image processing, and a fast mode that runs at roughly 2.5x output speed. Opus 4.8 is also markedly less likely than Opus 4.7 to leave undetected flaws in generated code, and is the recommended migration target for the deprecated Claude Opus 4 and Claude Sonnet 4.
MetricValue
Parameter CountUnknown
Mixture of ExpertsUnknown
Context Length1,000,000 tokens
Max Output128,000 tokens
MultilingualYes
Quantized*Unknown
*Quantization is specific to the inference provider and the model may be offered with different quantization levels by other providers.

​
Example request

Use the Workbench as a request builder: configure parameters for this model in the UI, then open the API tab to copy the exact cURL or Python call.
curl -X POST https://hub.oxen.ai/api/ai/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OXEN_API_KEY" \
  -d '{
  "model": "claude-opus-4-8",
  "messages": [
    {
      "role": "user",
      "content": "Hello, what can you do?"
    }
  ]
}'

​
Fetch model details

The models endpoint returns the full model object, including its json_request_schema. 
curl -H "Authorization: Bearer $OXEN_API_KEY" https://hub.oxen.ai/api/ai/models/claude-opus-4-8

​
Request parameters

This model follows the standard OpenAI chat completions request body. See the chat completions reference for the full parameter list.