Client tool compatibility
Check the matrix for the tool types and protocol your client uses.
Responses-native models
- GPT-5.4openai/gpt-5.4
- GPT-5.6 Solopenai/gpt-5.6-sol
- GPT-6 Astraopenai/gpt-6-astra
- GPT-6 Lunaopenai/gpt-6-luna
- GPT-6 Solopenai/gpt-6-sol
- GPT-6.1 Solopenai/gpt-6.1-sol
- Qwen: Qwen3.6 Flashqwen/qwen3.6-flash
- Qwen: Qwen3.6 Plusqwen/qwen3.6-plus
- Qwen: Qwen3.7 Flashqwen/qwen3.7-flash
- Qwen: Qwen3.8 Maxqwen/qwen3.8-max
- Qwen:Qwen3.7 Plusqwen/qwen3.7-plus
- Qwen:Qwen3.8 flashqwen/qwen3.8-flash
Tool type compatibility matrix
| Tool type | When routed to a Responses-native route | Otherwise |
|---|---|---|
function | Supported | Supported |
custom (e.g. Codex exec) | Supported | Adapted (Gemini routes pending) |
local_shell | Supported | Adapted |
namespace (tool groups) | Supported (flattened) | Same |
tool_search · execution:"client" | Supported | Adapted for OpenAI/Anthropic protocols (Gemini unverified) |
tool_search · server-side search | Supported | Not supported: 400 with reason=server_executed |
shell (local) / apply_patch | Supported | Returns 400 with reason=not_adapted |
computer / computer_use_preview | Supported | Planned (needs multimodal tool output) |
mcp / web_search / file_search / code_interpreter / image_generation | Supported | Not supported: 400 with reason=server_executed |
Status reflects the current production release and is updated on deploy; rejected requests state the reason and available alternatives in the error message.
Anthropic output and thinking controls
For Messages requests routed to an Anthropic upstream, see the Anthropic field reference for the supported output_config and thinking shapes.