Grok 4 Fast (Reasoning)

Grok 4 Fast (Reasoning) is a model from SpaceXAI, released September 19, 2025. Its best result is #53 of 109 on MedScribe.

Release Date: Sep 19, 2025

Developer SpaceXAI  🇺🇸
Context Window 2M
Max Output Tokens 2M
Token Costs (in/out) $0.20/0.50
Weights Private
Input Modalities

Accuracy

59.07 %

Avg. Cost (In/Out)

$ 0.20 / $ 0.50

Latency

3 min 0 s

Vals Index
BenchmarksAccuracyRankings

37.38%

±1.94
76/107

81.63%

±2.14
53/109

42.09%

±0.95
91/98

75.70%

±0.76
13/145

0.00%

±0.00
131/132

85.35%

±1.79
49/138

78.97%

±1.10
76/143

80.60%

±0.45
85/151

79.70%

±0.39
95/138

72.78%

±1.07
61/93

45.40%

±2.23
84/88

29.21%

±4.85
46/67
Proprietary BenchmarksAcademic BenchmarksIndustry Partners
Vals
Default Provider : SpaceXAI
Temperature: 0.7
Top P: 0.95
Top K: Default
Max Output Tokens: 2,000,000

Updates

Sep 20, 2025

We evaluated Grok 4 Fast (Reasoning) and Grok 4 Fast (Non-Reasoning), and found the following:

  • Grok 4 Fast (Reasoning) delivers impressive latency and pricing, comparable to GPT-5 Nano. For example, on Corp Fin, it was half the latency and 10% of the cost compared to Grok 4.
  • The Reasoning model significantly outperforms the Non-Reasoning variant. While Grok 4 Fast (Non-Reasoning) did not place in the top 10 on any benchmark, Grok Fast (Reasoning) achieved #3 on both AIME and CorpFin.
  • The results highlight the growing performance gap between reasoning and non-reasoning models, particularly on complex benchmarks.

Overall, this is a very performant model from the SpaceXAI team for its price and latency.