Model comparison
DeepSeek Coder 33B vs Grok 3
Grok 3 has enough public results to be ranked (#157); DeepSeek Coder 33B does not yet, so treat this comparison as directional.
Last verified . 1 shared benchmarks.
DeepSeekSummary
- They share 1 benchmark with published results for both. DeepSeek Coder 33B scores higher in 0 categories and Grok 3 in 1 category; one gap is clear of the uncertainty.
- The widest gap is in coding, where Grok 3 leads 41.9 to 38.0.
- DeepSeek Coder 33B has downloadable open weights; the other is API-only.
Side by side
| DeepSeek Coder 33B | Grok 3 | |
|---|---|---|
| Provider | DeepSeek | xAI |
| Noometry Index | 38.9 | 39.9 |
| Released | 2023-11-02 | 2025-04-09 |
| Weights | Open | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 9 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 3 leads
DeepSeek Coder 33B: 38.0 (#184), Grok 3: 41.9 (#115)
| Benchmark | DeepSeek Coder 33B | Grok 3 |
|---|---|---|
| Aider Polyglot | — | 53.3% |
| WeirdML | — | 37.2% |
| BigCodeBench Instruct | 42% | — |
| LMArena Coding | — | 1432 |
| BigCodeBench Complete | 51.1% | — |
| HumanEval+ | 75% | — |
| MBPP+ | 70.1% | — |
Agentic & Tool Use Not comparable
DeepSeek Coder 33B: —, Grok 3: 30.5 (#76)
| Benchmark | DeepSeek Coder 33B | Grok 3 |
|---|---|---|
| BALROG | — | 29.5% |
Reasoning Not comparable
DeepSeek Coder 33B: —, Grok 3: 13.7 (#333)
| Benchmark | DeepSeek Coder 33B | Grok 3 |
|---|---|---|
| Epoch Capabilities Index | 96.32 | 138.33 |
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 36.1% |
| Kagi LLM Benchmark | — | 61.3% |
| ARC-AGI-1 | — | 5.5% |
| LMArena Hard Prompts | — | 1434 |
| WinoGrande | 62% | — |
Math Not comparable
DeepSeek Coder 33B: —, Grok 3: 38.0 (#145)
| Benchmark | DeepSeek Coder 33B | Grok 3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 55.6% |
| Omni-MATH | — | 46.4% |
| LMArena Math | — | 1391 |
| MATH Level 5 | — | 88.7% |
| FrontierMath (Feb 2025 set) | — | 3.8% |
| FrontierMath Tier 4 (v1) | — | 0% |
| GSM8K | 35.4% | — |
Knowledge Not comparable
DeepSeek Coder 33B: —, Grok 3: 46.2 (#82)
| Benchmark | DeepSeek Coder 33B | Grok 3 |
|---|---|---|
| GPQA Diamond | — | 75.8% |
| MMLU-Pro | — | 78.8% |
| Confabulations | — | 14.2% |
| Vectara Hallucination Rate | — | 5.8% |
| GPQA (HELM) | — | 65% |
| LMArena Expert | — | 1421 |
| ARC (AI2) Challenge | 42.2% | — |
| MMLU | 39.4% | — |
Multilingual Not comparable
DeepSeek Coder 33B: —, Grok 3: 52.3 (#87)
| Benchmark | DeepSeek Coder 33B | Grok 3 |
|---|---|---|
| LMArena Non-English | — | 1410 |
| LMArena Chinese | — | 1448 |
| LMArena French | — | 1460 |
| LMArena German | — | 1431 |
| LMArena Japanese | — | 1387 |
| LMArena Korean | — | 1373 |
| LMArena Russian | — | 1416 |
| LMArena Spanish | — | 1417 |
Instruction Following Not comparable
DeepSeek Coder 33B: —, Grok 3: 75.0 (#73)
| Benchmark | DeepSeek Coder 33B | Grok 3 |
|---|---|---|
| IFEval | — | 88.4% |
| LMArena Instruction Following | — | 1409 |
Long Context Not comparable
DeepSeek Coder 33B: —, Grok 3: 38.7 (#192)
| Benchmark | DeepSeek Coder 33B | Grok 3 |
|---|---|---|
| Fiction.LiveBench | — | 58.3% |
| LMArena Longer Query | — | 1439 |
Writing & Preference Not comparable
DeepSeek Coder 33B: —, Grok 3: 55.8 (#141)
| Benchmark | DeepSeek Coder 33B | Grok 3 |
|---|---|---|
| LMArena Text | — | 1426 |
| LMArena Creative Writing | — | 1414 |
| Short-Story Creative Writing | — | 76.4% |
| EQ-Bench Creative Writing | — | 1186 |
| WildBench | — | 84.9% |
| LMArena Multi-Turn | — | 1425 |
Frequently asked questions
Is DeepSeek Coder 33B better than Grok 3?
Grok 3 has enough public results to be ranked (#157); DeepSeek Coder 33B does not yet, so treat this comparison as directional.
Is DeepSeek Coder 33B or Grok 3 better for coding?
Grok 3 scores higher on coding benchmarks: 41.9 versus 38.0 in the Noometry coding category.
How many benchmarks do DeepSeek Coder 33B and Grok 3 share?
1 benchmark has published results for both models. DeepSeek Coder 33B has 9 scored results on Noometry and Grok 3 has 40.