Text and chat

Live

OpenAI-compatible chat completions for text and reasoning workloads.

(opens in new tab)Last verified: 2026-08-11

Request model

Send an ordered messages array and a live text model ID. temperature and max_tokens are supported. Grok 4.5 and 4.6 also accept OpenAI-compatible tools, tool_choice, parallel_tool_calls, and response_format. Grok 4.5 supports reasoning_effort as low, medium, or high; Grok 4.6 also supports xhigh. Omit stream for a JSON completion, or set stream: true for OpenAI-compatible server-sent events; stream_options.include_usage is accepted. Compatibility clients may send store: false.

bash
curl https://api.modelrush.ai/v1/chat/completions \
  -H "Authorization: Bearer $MODELRUSH_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "modelrush/grok-4.6",
    "messages": [{"role": "user", "content": "Return exactly: ModelRush connected"}]
  }'

bash
curl https://api.modelrush.ai/v1/chat/completions \
  -H "Authorization: Bearer $MODELRUSH_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"modelrush/grok-4.6","messages":[{"role":"user","content":"Stream a short answer"}],"stream":true}'

Response contract

The response follows the Chat Completions shape with id, model, choices, and usage. Persist the response ID for support and reconciliation.

Limits

ModelRush does not impose a global message-count limit and never silently truncates, merges, deletes, or reorders conversation history. Capacity is determined by the selected model's context window, the published request-body limit, and any provider-specific limit declared for that model operation. Read versioned limits from GET /models; do not copy one model's context window or output cap into another integration. Oversized bodies return request_too_large; context overflow returns context_length_exceeded.

PreviousModel API overview

NextImage generation