Grok 4 Fast (Reasoning)
xAIGrokProprietary
Description
Pushing the Frontier of Cost-Efficient Intelligence. Grok-4 Fast Reasoning is a high-speed variant of Grok-4 optimized for faster inference while maintaining strong reasoning capabilities through thinking tokens.
Release Date
2025-09-19
Parameters
—
Context Length
—
Modalities
image, text
Capability Radar
43
general
75
coding
89
reasoning
57
scienceest.
85
agents
80
multimodal
Science uses a reasoning proxy when dedicated science benchmarks are unavailable.
Rankings
| Domain | #Rank | Score | Source |
|---|---|---|---|
| Code Ranking | 89 | 69.0 | AA |
| General Ranking | 122 | 61.0 | AA |
| Math Reasoning | 38 | 91.0 | AA |
| Science | 85 | 64.0 | AA |
Benchmark Scores (LLM Stats)
No benchmark data available
AA Evaluation Indices
Math Index89.7
Intelligence Index27.4
Aime 250.9
Mmlu Pro0.8
Gpqa0.8
Livecodebench0.8
Tau20.7
Lcr0.6
Ifbench0.5
Scicode0.4
Terminalbench Hard0.2
Hle0.2
LLM Stats Category Scores
No category score data available
Pricing
Input Price$0.2 / 1M tokens
Output Price$0.5 / 1M tokens
Blended Price (3:1)$0.275 / 1M tokens
Speed
Tokens/sec0.0
Time to First Token0.00s
Time to Answer0.00s
Provider Price Ranking
Provider Price Ranking
9 providers
Cheapest: xAIMost Expensive: Azure
ProviderInputOutput
1xAICheapest
$0
$0
2Jiekou.AI
$0.18
$0.45
3Helicone
$0.2
$0.5
4302.AI
$0.2
$0.5
5Abacus
$0.2
$0.5
6Poe
$0.2
$0.5
7Azure Cognitive Services
$0.2
$0.5
8LLM Gateway
$0.2
$0.5
9Azure
$0.2
$0.5
Compare pricing across different API providers for this model.