Build

Thinking and reasoning

Choose reasoning effort or a thinking budget, and understand the token and cost impact.

Choose a control

Check the model's reasoning_options in the public catalog. It describes an effort list, a token budget, or a toggle. Use the model's listed values and limits.

Control	Chat Completions field
Effort level	`reasoning: {"effort":"high"}` using one of the model's listed levels
Alternative effort syntax	`reasoning_effort: "high"`; do not also send `reasoning`
Reasoning token budget	`reasoning: {"max_tokens":2048}` for a budget-based model
Disable reasoning	`reasoning: {"effort":"none"}` only when the model supports disabling it
Hide returned reasoning	`reasoning: {"effort":"high","exclude":true}` where supported; this does not disable reasoning or its cost
Model default	Omit the reasoning fields

Effort names can include none, minimal, low, medium, high, xhigh, and max. Each model supports its own subset. More effort can increase latency and cost; it does not guarantee a better answer.

Set your key

Create an API key, save its recovery link, and add credits. Replace the placeholder below and run it in your terminal before running the examples on this page.

Set your key

export MINIROUTER_KEY='paste-your-api-key-here'

Examples use paid models. For free requests, follow the Auto Free guide.

The model IDs below are examples. To use another model, check its supported APIs and reasoning controls in the catalog.

Chat Completions

Set effort with reasoning.effort and leave room in max_completion_tokens for thinking and the answer.

Chat Completions

curl https://api.minirouter.sh/v1/chat/completions \
  -H "Authorization: Bearer $MINIROUTER_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "openai/gpt-6-astra",
  "messages": [
    {
      "role": "user",
      "content": "What is the smallest positive integer divisible by every integer from 1 through 10? Explain briefly."
    }
  ],
  "max_completion_tokens": 4096,
  "reasoning": {
    "effort": "high"
  }
}'

The final answer is in choices[0].message.content. Reasoning text, when returned, is separate from the answer and varies by model.

Responses

Use input, reasoning.effort, and max_output_tokens.

Responses

curl https://api.minirouter.sh/v1/responses \
  -H "Authorization: Bearer $MINIROUTER_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "openai/gpt-6-astra",
  "store": false,
  "input": "What is the smallest positive integer divisible by every integer from 1 through 10? Explain briefly.",
  "max_output_tokens": 4096,
  "reasoning": {
    "effort": "high"
  }
}'

Read the answer from output message items with output_text content. This example returns JSON. Use Chat Completions or Messages for streaming.

Messages

For adaptive thinking, use thinking.type: "adaptive" and output_config.effort. Include the Messages version header.

Messages

curl https://api.minirouter.sh/v1/messages \
  -H "Authorization: Bearer $MINIROUTER_KEY" \
  -H "Content-Type: application/json" \
  -H "anthropic-version: 2023-06-01" \
  -d '{
  "model": "anthropic/claude-haiku-5.5",
  "max_tokens": 4096,
  "thinking": {
    "type": "adaptive"
  },
  "output_config": {
    "effort": "high"
  },
  "messages": [
    {
      "role": "user",
      "content": "Compare two approaches to caching API responses."
    }
  ]
}'

Read text blocks in content for the answer; thinking blocks are separate.

For a model with a fixed thinking budget, use thinking: {"type":"enabled","budget_tokens":2048} instead of adaptive thinking. The budget must be below max_tokens and within the model's supported range. Do not add budget_tokens to adaptive thinking.

To disable thinking where supported, use thinking: {"type":"disabled"} and omit output_config.effort. Messages does not accept none as an effort.

Read streamed reasoning

On Chat Completions streams, a model may return delta.reasoning_content, delta.reasoning, or delta.reasoning_details separately from delta.content. Messages streams can include thinking blocks and thinking_delta events. Not every model exposes its reasoning.

Preserve provider-issued thinking blocks and signatures when continuing a Messages conversation. Do not replace them with a reconstruction of the text. See streaming for event handling and cost.

Troubleshooting

Symptom	What to check
No thinking control in your client	Refresh its model list, check its setup guide, or send an API example above. Some clients need reasoning enabled in their model configuration.
No visible thinking text	Thinking can be hidden by the model or client. Check reasoning-token usage when returned; an absent thinking panel does not establish whether thinking ran.
Empty or cut-off answer	Increase the endpoint's output-token limit. Thinking and the answer share that allowance. Check for Chat Completions `finish_reason: "length"`, Responses `status: "incomplete"`, or Messages `stop_reason: "max_tokens"`.
Unsupported effort or budget	Check the model's `reasoning_options` for allowed effort values, token budgets, and support for disabling reasoning.
Spend-limit or balance error	Check available credits and the key's spend limit. Admission reserves a worst-case hold before sending the request, which can exceed its eventual charge.

Tokens and cost

Reasoning tokens can count toward billed output even when hidden. Allow room for both reasoning and the visible answer. Reasoning requests may require a larger temporary balance hold than a request without reasoning; reducing the visible output limit alone may not reduce that hold. Check the final cost in usage.cost or Request logs.