Model comparison
DeepSeek-V3.1-Terminus vs Grok 3
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 39.9 on the Noometry Index.
Last verified . 11 shared benchmarks.
DeepSeekSummary
- They share 11 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 5 categories and Grok 3 in 2 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek-V3.1-Terminus leads 26.4 to 13.7.
- DeepSeek-V3.1-Terminus has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3.1-Terminus | Grok 3 | |
|---|---|---|
| Provider | DeepSeek | xAI |
| Noometry Index | 43.1 | 39.9 |
| Released | 2025-09-22 | 2025-04-09 |
| Weights | Open | Proprietary |
| Context window | 164K | — |
| Max output | 147K | — |
| Input $ / M tokens | $0.27 | — |
| Output $ / M tokens | $1 | — |
| Results tracked | 16 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
DeepSeek-V3.1-Terminus: 42.0 (#113), Grok 3: 41.9 (#115)
| Benchmark | DeepSeek-V3.1-Terminus | Grok 3 |
|---|---|---|
| LMArena Coding | 1426 | 1432 |
| Aider Polyglot | — | 53.3% |
| SciCode | 40.6% | — |
| WeirdML | — | 37.2% |
| ALE-Bench | 745.17 | — |
Agentic & Tool Use Not comparable
DeepSeek-V3.1-Terminus: —, Grok 3: 30.5 (#76)
| Benchmark | DeepSeek-V3.1-Terminus | Grok 3 |
|---|---|---|
| BALROG | — | 29.5% |
Reasoning DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 26.4 (#133), Grok 3: 13.7 (#333)
| Benchmark | DeepSeek-V3.1-Terminus | Grok 3 |
|---|---|---|
| Kagi LLM Benchmark | 57.4% | 61.3% |
| LMArena Hard Prompts | 1426 | 1434 |
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 36.1% |
| ARC-AGI-1 | — | 5.5% |
| CritPt | 1.7% | — |
| DTBench | 81.3% | — |
| LMCA | 28.6% | — |
| Epoch Capabilities Index | — | 138.33 |
Math Too close to call
DeepSeek-V3.1-Terminus: 38.5 (#137), Grok 3: 38.0 (#145)
| Benchmark | DeepSeek-V3.1-Terminus | Grok 3 |
|---|---|---|
| LMArena Math | 1402 | 1391 |
| OTIS Mock AIME 2024-2025 | — | 55.6% |
| Omni-MATH | — | 46.4% |
| MATH Level 5 | — | 88.7% |
| FrontierMath (Feb 2025 set) | — | 3.8% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Not comparable
DeepSeek-V3.1-Terminus: —, Grok 3: 46.2 (#82)
| Benchmark | DeepSeek-V3.1-Terminus | Grok 3 |
|---|---|---|
| GPQA Diamond | — | 75.8% |
| MMLU-Pro | — | 78.8% |
| Confabulations | — | 14.2% |
| Vectara Hallucination Rate | — | 5.8% |
| GPQA (HELM) | — | 65% |
| LMArena Expert | — | 1421 |
Multilingual Too close to call
DeepSeek-V3.1-Terminus: 52.1 (#92), Grok 3: 52.3 (#87)
| Benchmark | DeepSeek-V3.1-Terminus | Grok 3 |
|---|---|---|
| LMArena Non-English | 1407 | 1410 |
| LMArena Russian | 1436 | 1416 |
| LMArena Chinese | — | 1448 |
| LMArena French | — | 1460 |
| LMArena German | — | 1431 |
| LMArena Japanese | — | 1387 |
| LMArena Korean | — | 1373 |
| LMArena Spanish | — | 1417 |
Instruction Following Grok 3 leads
DeepSeek-V3.1-Terminus: 74.0 (#106), Grok 3: 75.0 (#73)
| Benchmark | DeepSeek-V3.1-Terminus | Grok 3 |
|---|---|---|
| LMArena Instruction Following | 1404 | 1409 |
| IFEval | — | 88.4% |
Long Context DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 43.4 (#97), Grok 3: 38.7 (#192)
| Benchmark | DeepSeek-V3.1-Terminus | Grok 3 |
|---|---|---|
| LMArena Longer Query | 1421 | 1439 |
| Fiction.LiveBench | — | 58.3% |
Writing & Preference DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 61.0 (#92), Grok 3: 55.8 (#141)
| Benchmark | DeepSeek-V3.1-Terminus | Grok 3 |
|---|---|---|
| LMArena Text | 1419 | 1426 |
| LMArena Creative Writing | 1403 | 1414 |
| LMArena Multi-Turn | 1411 | 1425 |
| Short-Story Creative Writing | — | 76.4% |
| EQ-Bench Creative Writing | — | 1186 |
| WildBench | — | 84.9% |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than Grok 3?
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 39.9 on the Noometry Index.
Is DeepSeek-V3.1-Terminus or Grok 3 better for coding?
They score almost the same on coding (42.0 vs 41.9); test both on your own repository before choosing.
How many benchmarks do DeepSeek-V3.1-Terminus and Grok 3 share?
11 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Grok 3 has 40.