Up to four models
DeepSeek-V3 vs Grok-2 (Dec 2024)
On the index (GPT-4 = 100), DeepSeek-V3 scores 119 and Grok-2 (Dec 2024) 113. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
119103–135
WeightsOpen
ReleasedDec 2024
Price in / out$0.26 / $1.03
Context164K
Core tests taken2 of 9
Grok-2 (Dec 2024)
11397–129
WeightsClosed
ReleasedDec 2024
Price in / out–
Context–
Core tests taken1 of 9
DeepSeek-V3 and Grok-2 (Dec 2024) are too close to call (6 points apart, within the uncertainty), ahead on 1 of 1 shared core tests.
Test by testthird-party results
Every result as a table
| Test | DeepSeek-V3 | Grok-2 (Dec 2024) |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 56.5% | 53.8% |
| SimpleBench | 18.9% | 22.7% |
| Maths | ||
| AIME-style maths (OTIS mock) | 15.8% | 11.5% |
| MATH (level 5) | 64.8% | 63.5% |
| FrontierMath (2025 set) | 1.7% | 0.7% |
Task length · METR
28 sec
Tests at least two of these models have taken. Bold rows are the core tests.