Up to four models
Grok Build 0.1 vs DeepSeek-V3.2
On the index (GPT-4 = 100), Grok Build 0.1 scores 141 and DeepSeek-V3.2 141. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
Grok Build 0.1
141114–169
WeightsClosed
ReleasedMay 2026
Price in / out$1.0 / $2.0
Context256K
Core tests taken0 of 9
141125–157
WeightsOpen
ReleasedDec 2025
Price in / out$0.26 / $0.38
Context164K
Core tests taken2 of 9
Grok Build 0.1 and DeepSeek-V3.2 are too close to call (0 points apart, within the uncertainty), at 4.3× the price.
Test by testthird-party results
Grok Build 0.1
DeepSeek-V3.2
These models share no third-party tests yet.
Tests at least two of these models have taken. Bold rows are the core tests.