Model comparison
DeepSeek-R1 vs Grok-2 (Dec 2024)
DeepSeek-R1 is the stronger model overall, scoring 42.3 to 33.7 on the Noometry Index.
Last verified . 32 shared benchmarks.
DeepSeekSummary
- They share 32 benchmarks with published results for both. DeepSeek-R1 scores higher in 8 categories and Grok-2 (Dec 2024) in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek-R1 leads 43.8 to 20.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 66.4% for DeepSeek-R1 and 11.5% for Grok-2 (Dec 2024).
Side by side
| DeepSeek-R1 | Grok-2 (Dec 2024) | |
|---|---|---|
| Provider | DeepSeek | xAI |
| Noometry Index | 42.3 | 33.7 |
| Released | 2025-01-20 | 2024-08-13 |
| Weights | Proprietary | Proprietary |
| Context window | 164K | — |
| Max output | 64K | — |
| Input $ / M tokens | $0.50 | — |
| Output $ / M tokens | $2.15 | — |
| Results tracked | 52 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-R1 leads
DeepSeek-R1: 46.3 (#68), Grok-2 (Dec 2024): 33.3 (#258)
| Benchmark | DeepSeek-R1 | Grok-2 (Dec 2024) |
|---|---|---|
| WeirdML | 41.6% | 22.2% |
| LiveBench Coding | 66.7% | 46.4% |
| LMArena Coding | 1427 | 1287 |
| Aider Polyglot | 71.4% | — |
| SciCode | 35.7% | — |
| ALE-Bench | 804.12 | — |
| AlgoTune | 1.7 | — |
Agentic & Tool Use Not comparable
DeepSeek-R1: 30.7 (#75), Grok-2 (Dec 2024): —
| Benchmark | DeepSeek-R1 | Grok-2 (Dec 2024) |
|---|---|---|
| DeepResearch Bench | 35.1% | — |
| BALROG | 34.9% | — |
| METR Time Horizons | 53.8% | — |
Reasoning DeepSeek-R1 leads
DeepSeek-R1: 18.6 (#278), Grok-2 (Dec 2024): 16.9 (#299)
| Benchmark | DeepSeek-R1 | Grok-2 (Dec 2024) |
|---|---|---|
| SimpleBench | 40.8% | 22.7% |
| LiveBench Reasoning | 83.2% | 54.8% |
| LMArena Hard Prompts | 1416 | 1272 |
| LiveBench Data Analysis | 69.8% | 54.5% |
| Epoch Capabilities Index | 141.29 | 130.48 |
| LiveBench | 71.6% | 54.3% |
| ARC-AGI-2 | 1.3% | — |
| Kagi LLM Benchmark | 69.4% | — |
| ARC-AGI-1 | 21.2% | — |
| CritPt | 1.1% | — |
| DTBench | — | 65.2% |
| ForecastBench | 60 | — |
Math DeepSeek-R1 leads
DeepSeek-R1: 43.8 (#79), Grok-2 (Dec 2024): 20.8 (#284)
| Benchmark | DeepSeek-R1 | Grok-2 (Dec 2024) |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.4% | 11.5% |
| LiveBench Math | 80.7% | 54.9% |
| LMArena Math | 1400 | 1283 |
| MATH Level 5 | 96.6% | 63.5% |
| Omni-MATH | 42.4% | — |
| FrontierMath (Feb 2025 set) | — | 0.7% |
Knowledge DeepSeek-R1 leads
DeepSeek-R1: 44.5 (#87), Grok-2 (Dec 2024): 29.8 (#233)
| Benchmark | DeepSeek-R1 | Grok-2 (Dec 2024) |
|---|---|---|
| GPQA Diamond | 76.3% | 53.8% |
| Confabulations | 12.7% | 20.1% |
| LMArena Expert | 1394 | 1254 |
| MMLU-Pro | 79.3% | — |
| Vectara Hallucination Rate | 11.3% | — |
| GPQA (HELM) | 66.6% | — |
Multilingual DeepSeek-R1 leads
DeepSeek-R1: 52.4 (#85), Grok-2 (Dec 2024): 43.1 (#188)
| Benchmark | DeepSeek-R1 | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Non-English | 1412 | 1282 |
| LMArena Chinese | 1442 | 1289 |
| LMArena French | 1417 | 1318 |
| LMArena German | 1404 | 1287 |
| LMArena Japanese | 1391 | 1244 |
| LMArena Korean | 1360 | 1237 |
| LMArena Russian | 1423 | 1286 |
| LMArena Spanish | 1411 | 1281 |
Instruction Following DeepSeek-R1 leads
DeepSeek-R1: 72.0 (#143), Grok-2 (Dec 2024): 66.9 (#202)
| Benchmark | DeepSeek-R1 | Grok-2 (Dec 2024) |
|---|---|---|
| LiveBench Instruction Following | 80.5% | 69.6% |
| LMArena Instruction Following | 1382 | 1270 |
| IFEval | 78.4% | — |
Long Context DeepSeek-R1 leads
DeepSeek-R1: 45.4 (#36), Grok-2 (Dec 2024): 38.8 (#190)
| Benchmark | DeepSeek-R1 | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Longer Query | 1391 | 1276 |
| Fiction.LiveBench | 75% | — |
Writing & Preference DeepSeek-R1 leads
DeepSeek-R1: 61.4 (#88), Grok-2 (Dec 2024): 48.6 (#198)
| Benchmark | DeepSeek-R1 | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Text | 1428 | 1305 |
| LMArena Creative Writing | 1405 | 1284 |
| Short-Story Creative Writing | 83% | 63.6% |
| LMArena Multi-Turn | 1405 | 1290 |
| LiveBench Language | 48.5% | 45.6% |
| EQ-Bench Creative Writing | 1500 | — |
| WildBench | 82.8% | — |
Frequently asked questions
Is DeepSeek-R1 better than Grok-2 (Dec 2024)?
DeepSeek-R1 is the stronger model overall, scoring 42.3 to 33.7 on the Noometry Index.
Is DeepSeek-R1 or Grok-2 (Dec 2024) better for coding?
DeepSeek-R1 scores higher on coding benchmarks: 46.3 versus 33.3 in the Noometry coding category.
How many benchmarks do DeepSeek-R1 and Grok-2 (Dec 2024) share?
32 benchmarks have published results for both models. DeepSeek-R1 has 52 scored results on Noometry and Grok-2 (Dec 2024) has 34.