1. 7API
  2. Docs
  3. Reasoning effort

Reasoning effort

Most models in the catalog reason before they answer. Higher effort usually means better answers on hard problems, with more output tokens and a longer wait before the first visible token. Reasoning tokens are billed as output.

Parameters

Send whichever parameter your client already uses. The gateway converts it to what the model expects.

ParameterFormatExample
reasoning_effortChat Completions"reasoning_effort": "high"
reasoning.effortResponses"reasoning": {"effort": "high"}
output_config.effortAnthropic Messages (used by Claude Code)"output_config": {"effort": "high"}
thinkingAnthropic Messages{"type": "adaptive"}. {"type": "enabled", "budget_tokens": N} is accepted on claude-opus-4-6 only.

Levels by model family

ModelsAccepted levelsNotes
Claude (Opus, Sonnet, Haiku)low, medium, high, xhigh, maxclaude-opus-4-6 has no xhigh.
GPT-6 / GPT-5.6none, low, medium, high, xhigh, maxgpt-5.5 stops at xhigh. On gpt-6.1-sol, max becomes xhigh and none becomes low.
GLM 5.3, GLM 5.3 Flashlow, high, xhigh, maxOther values become high.
GLM 5.3 Abliteratedhigh, xhigh, maxThe minimum and default is high. Thinking can't be turned off.
GLM 5.3 Uncensored, GLM 5.3 Flash UncensoredmaxAlways maximum effort
DeepSeek V4 / V4.1low, medium, high
DeepSeek V4.1 Flash 1017 Abliteratednone, low, high, maxDefaults to high
Kimi K3, Kimi K3.1 Prolow, high, maxDefaults to max. medium becomes high, xhigh becomes max, and none, off and minimal become low. Any other value returns 400 invalid_reasoning_effort.

GET /v1/models reports supported_reasoning_efforts and default_reasoning_effort for models with a fixed scale. That lets clients such as OpenCode and Oh My Pi build their effort pickers automatically.

Output budget

Reasoning and the final answer share the request's output limit (max_tokens or an equivalent). At high effort, set the limit generously. The GLM abliterated models default to a 130,000-token output budget for this reason.