Model detail
xAI: Grok 4 Fast Reasoning
Grok 4 Fast is xAI's latest multimodal model with SOTA cost-efficiency and a 2M token context window. It comes in two flavors: non-reasoning and reasoning. Read more about the model on xAI's [news post](http://x.ai/news/grok-4-fast). Reasoning can be enabled/disabled using the `reasoning` `enabled` parameter in the API. [Learn more in our docs](https://openrouter.ai/docs/use-cases/reasoning-tokens#controlling-reasoning-tokens)
Model specs
- Model ID
-
grok-4-fast-reasoning
- Context length
- 2M
- Max output
- 2M
- I/O modalities
- Released
- 2025-09
02
API endpoints
One MixRoute gateway, OpenAI-compatible
-
OpenAI-compatible
/v1/chat/completionsPOST
03
Pricing
Flat rate, unit: /1M Tokens
Input
$0.2000 /1M Tokens
Completion
$0.5000 /1M Tokens
Cache read
$0.0500 /1M Tokens
05
Selection summary
Quickly judge whether it fits your workload
What is this model good for?
Grok 4 Fast is xAI's latest multimodal model with SOTA cost-efficiency
Grok 4 Fast is xAI's latest multimodal model with SOTA cost-efficiency and a 2M token context window. It comes in two flavors: non-reasoning and reasoning. Read more about the model on xAI's news post.
What should you check before using it?
2,000,000 context, 2,000,000 output tokens
xAI model that supports reasoning, tool calling and structured output. Knowledge cutoff: 2025-09. Released 2025-09-19.
Why use it through MixRoute?
xAI: Grok 4 Fast Reasoning keeps its 2,000,000-token context on one route
The full 2,000,000-token context stays reachable on the same route. MixRoute serves xAI: Grok 4 Fast Reasoning on the OpenAI-compatible endpoint /v1/chat/completions; point your existing client at https://api.mixroute.ai/v1 with the model ID grok-4-fast-reasoning and re-run your integration tests.
07
Capabilities & compatibility
Unconfirmed capabilities stay "Unknown," no guesses
Tool calling
xAI: Grok 4 Fast Reasoning supports function and tool calls.
tools definition; the response returns tool_calls to execute and feed results back.
Structured output
xAI: Grok 4 Fast Reasoning can produce JSON-formatted output on demand.
response_format: { type: "json_schema" } forces valid JSON output.
Streaming
xAI: Grok 4 Fast Reasoning supports streamed responses.
stream: true streams tokens incrementally, so chat UIs render as they arrive.
Prompt caching
xAI: Grok 4 Fast Reasoning supports prompt caching under supported routing conditions.
Vision input
xAI: Grok 4 Fast Reasoning can process image inputs alongside text.
image_url content part alongside text messages.
OpenAI SDK
Call xAI: Grok 4 Fast Reasoning through MixRoute's OpenAI-compatible endpoint.
base_url to https://api.mixroute.ai/v1; everything else follows the OpenAI style.
08
Token cost estimator
Live estimate from this page's pricing, not an actual bill
Share of the same prefix read repeatedly, up to 100%
09
FAQ
Recorded results for xAI: Grok 4 Fast Reasoning: aa_intelligence_index 28.6; arena_text 81.5. Benchmark positions are a coarse filter only — run your own prompts against the candidate models before you commit, because ranking shifts with task shape and prompt style.
One endpoint, a testable decision
Test this model and alternate routes with the same request format
Start from a real workload, then let quality, total cost, and failure conditions decide whether to send production traffic.