Compare · ModelsLive · 2 picked · head to head
Gemini 3 Flash Preview vs Grok 4 Fast
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Gemini 3 Flash Preview wins on 4/4 benchmarks
Gemini 3 Flash Preview wins 4 of 4 shared benchmarks. Leads in reasoning · general · coding.
Category leads
reasoning·Gemini 3 Flash Previewgeneral·Gemini 3 Flash Previewcoding·Gemini 3 Flash Preview
Hype vs Reality
Attention vs performance
Gemini 3 Flash Preview
#135 by perf·#10 by attention
Grok 4 Fast
#105 by perf·#14 by attention
Best value
Gemini 3 Flash Preview
Gemini 3 Flash Preview
28.5 pts/$
$1.75/M
Grok 4 Fast
n/a
no price
Vendor risk
Who is behind the model
Google DeepMind
$3.53T·Tier 1
xAI
$250.0B·Tier 1
Head to head
4 benchmarks · 2 models
Gemini 3 Flash PreviewGrok 4 Fast
ARC-AGI
Gemini 3 Flash Preview leads by +36.2
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
Gemini 3 Flash Preview
84.7
Grok 4 Fast
48.5
ARC-AGI-2
Gemini 3 Flash Preview leads by +28.3
ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data.
Gemini 3 Flash Preview
33.6
Grok 4 Fast
5.3
Dtbench
Gemini 3 Flash Preview leads by +10.7
Gemini 3 Flash Preview
81.8
Grok 4 Fast
71.1
WeirdML
Gemini 3 Flash Preview leads by +18.7
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Gemini 3 Flash Preview
61.6
Grok 4 Fast
42.9
Full benchmark table
| Benchmark | Gemini 3 Flash Preview | Grok 4 Fast |
|---|---|---|
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 84.7 | 48.5 |
ARC-AGI-2 ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data. | 33.6 | 5.3 |
Dtbench | 81.8 | 71.1 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 61.6 | 42.9 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $0.50 | $3.00 | 1.0M tokens (~524 books) | $11.25 | |
| n/a | n/a | n/a | n/a |
People also compared