What this error means
Claude API 400 Bad Request — service_tier rewritten to deprecated 'speed' field is a Anthropic API failure pattern reported for developers trying to fix anthropic api 400 error when using service_tier through portkey gateway. Based on the imported evidence, treat this as a tool-specific troubleshooting page rather than a generic API error.
Why this happens
Official Portkey-AI/gateway GitHub issue. Portkey gateway translates OpenAI-style service_tier to Anthropic's deprecated 'speed' field per docs-core #869, but Anthropic's Messages API no longer accepts 'speed' — it uses service_tier natively. Result: any request with service_tier set returns HTTP 400. Priority Tier unreachable through gateway.
Common causes
- Developers using Portkey as an LLM gateway to route requests to the Anthropic API receive HTTP 400 errors because Portkey incorrectly translates OpenAI-style service_tier to Anthropic's deprecated 'speed' field. This makes Priority Tier (fast mode) completely unreachable through the gateway, blocking production workloads that depend on reduced latency.
- Official Portkey-AI/gateway GitHub issue. Portkey gateway translates OpenAI-style service_tier to Anthropic's deprecated 'speed' field per docs-core #869, but Anthropic's Messages API no longer accepts 'speed' — it uses service_tier natively. Result: any request with service_tier set returns HTTP 400. Priority Tier unreachable through gateway.
Quick fixes
- Confirm the exact error signature matches
Claude API 400 Bad Request — service_tier rewritten to deprecated 'speed' field. - Check the Anthropic API account, local tool state, and provider configuration involved in the failing workflow.
- Compare the failing environment with a known working setup, then change one configuration value at a time.
Platform/tool-specific checks
- Verify the command, editor, extension, or API client that produced the error.
- Compare local settings with CI, deployment, or editor-level settings when the error appears in only one environment.
- Avoid deleting credentials, local model data, or project settings until the failing scope is clear.
Step-by-step troubleshooting
- Capture the exact error message and the command, editor action, or request that triggered it.
- Check whether the failure is account/auth, quota/rate, model/provider, local runtime, or deployment configuration.
- Review the source evidence below and compare it with your environment.
- Apply one change at a time and rerun the smallest failing action.
- Keep the working fix documented for the team or deployment environment.
How to prevent it
- Keep provider/tool configuration documented.
- Record non-secret diagnostics such as tool version, provider name, model name, and command path.
- Add a lightweight check before CI or production workflows depend on the tool.