Model prices / xAI / grok-4-fast-reasoning
grok-4-fast-reasoning
xAI · released 2025-12-18
No price from the maker itself is on record for grok-4-fast-reasoning. The lowest of 5 providers is 302.AI: $0.2 per 1M input tokens and $0.5 per 1M output tokens (as of 2026-11-12). Context window: 2M tokens, up to 2M output tokens per request.
Price by providerUSD per 1M tokens
| Provider | Input | Output | Cached | Notes | |
|---|---|---|---|---|---|
3302.AILowestgrok-4-fast-reasoning | $0.2 | $0.5 | — | Lowest | Official docs |
HHeliconegrok-4-fast-reasoning | $0.2 | $0.5 | $0.05 | Official docs | |
JJiekou.AIDeprecatedgrok-4-fast-reasoning | $0.18 | $0.45 | — | Deprecated | Official docs |
 PoeDeprecatedxai/ | $0.2 | $0.5 | $0.05 | Deprecated | Official docs |
QQiniuDeprecatedx-ai/ | — | — | — | Deprecated | Official docs |
Providers are those listed by models.dev. This site has not vetted them one by one; each provider's own page has the binding price.
About
Fast Grok model for responsive chat, reasoning, and tool-assisted work
Description from models.dev.
Specs
- Type
- Text
- Context
- 2M
- Max output
- 2M
- Input
- Text, Image
- Output
- Text
- Reasoning
- Yes
- Tool calling
- Yes
- Structured output
- No
- Knowledge cutoff
- 2025-06
- Open weights
- No
- Released
- 2025-12-18
Source
Prices and specs come from models.dev (MIT licence), updated 2026-11-12. Each provider's name links to its official docs. A provider priced at 0 usually includes the model in a subscription; it is not necessarily a free API.