Grok 4 Fast
xai/grok-4-fastTwo million tokens of context at commodity prices, long-context economics, reset.
Model overview
Context window
2M
tokens
Input price
—
quoted on request
Output price
—
quoted on request
Weekly volume
—
tokens / week
Open weights
No
API only
Capabilities
4
of 11 tags
Description
Grok 4 Fast is a language model from xAI. Webparam does not stock this model yet and can source it on request — tell us what you need it for and we will come back with availability and a price. xAI’s stated focus is real-time-grounded frontier models. The weights are not published, so it is API-only.
Capabilities
- Tool calling
- Calls functions you define and returns their arguments as structured data, the basis of agents.
- JSON mode
- Constrains the response to valid JSON matching a schema you supply.
- Streaming
- Emits tokens as they are generated, so answers appear progressively rather than all at once.
- Long context
- Built for inputs well beyond the mainstream window, whole repositories, archives or corpora.
Strengths & weaknesses
Strengths
- 2M context at workhorse prices
- Fast
- Good tool calling
Weaknesses
- Depth trails Grok 4.1
- Recall softens at extreme depth
Pricing
| Rate | Price | Unit |
|---|---|---|
| Input | — quoted on request | — |
| Output | — quoted on request | — |
This model is not provisioned yet; the rate is quoted on request.
Context window
2Mtokens
covers prompt and response together, so a long input leaves less room for the answer.
Supported features
| Feature | Support |
|---|---|
| Tool calling | Supported |
| JSON mode | Supported |
| Streaming | Supported |
| Vision input | Not supported |
| Audio input | Not supported |
| Long context | Supported |
| Open weights | Not supported |
| Reasoning | Not supported |
| Fine-tunable | Not supported |
| Batch | Not supported |
| Caching | Not supported |
Example use cases
Long-context volume work
Archive Q&A
Cost-tiered routing
Comparison
Compare against anything →| Attribute | Grok 4 Fast | Llama 4 Scout | MiniMax Text 01 |
|---|---|---|---|
| Context window | 2M | 10M | 1M |
| Input price | R0.00 | R0.00 | R0.00 |
| Output price | — not charged | — not charged | — not charged |
| Tool calling | Supported | Supported | Supported |
| JSON mode | Supported | Not supported | Not supported |
| Vision input | Not supported | Not supported | Not supported |
| Open weights | Not supported | Supported | Supported |
| Prompt caching | Not supported | Not supported | Not supported |
Best value in each row is highlighted. Blank cells are facts the supplier does not publish, not zeroes.
Provider information
View xAI →xAI · San Francisco, United States · est. 2023
xAI built one of the largest training clusters in existence and iterated Grok to the frontier at unusual speed. Its models are distinguished by real-time grounding and enormous fast-inference context, Grok 4 Fast's 2M-token window at commodity prices reset long-context economics.
Recent releases
Documentation
Frequently asked questions
Input is billed at R0.00 per 1M tokens, and there is no separate output charge. Every figure on this page is an illustrative placeholder.
No. Grok 4 Fast is served through its provider's API only; the weights are not published, so self-hosting is not an option.
Prompt and response share one window: Grok 4 Fast accepts up to 2M tokens in total, so a long input leaves less room for the answer. The median across language models is 256K. It carries the long-context tag, meaning it is built for inputs well beyond the mainstream window.
Llama 4 Scout is the closest neighbour in the catalogue, with 10M of context at R0.00 per 1M tokens in. Put both in a comparison before wiring a fallback chain, so the trade-off is a decision rather than a default.