Model comparison
DeepSeek-V3.1 vs Tulu 3 (Tülu 3) 70B
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 33.0 on the Noometry Index.
Last verified . 11 shared benchmarks.
DeepSeek
Allen Institute for AI (Ai2)Summary
- They share 11 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 7 categories and Tulu 3 (Tülu 3) 70B in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek-V3.1 leads 38.9 to 14.2.
Side by side
| DeepSeek-V3.1 | Tulu 3 (Tülu 3) 70B | |
|---|---|---|
| Provider | DeepSeek | Allen Institute for AI (Ai2) |
| Noometry Index | 42.8 | 33.0 |
| Released | 2025-08-21 | 2024-11-21 |
| Weights | Open | Open |
| Context window | 164K | — |
| Max output | 8K | — |
| Input $ / M tokens | $0.25 | — |
| Output $ / M tokens | $0.95 | — |
| Results tracked | 27 | 14 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1 leads
DeepSeek-V3.1: 40.3 (#144), Tulu 3 (Tülu 3) 70B: 36.0 (#214)
| Benchmark | DeepSeek-V3.1 | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Coding | 1417 | 1235 |
| WeirdML | 38.4% | — |
Reasoning DeepSeek-V3.1 leads
DeepSeek-V3.1: 27.9 (#110), Tulu 3 (Tülu 3) 70B: 23.9 (#169)
| Benchmark | DeepSeek-V3.1 | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Hard Prompts | 1417 | 1220 |
| SimpleBench | 40% | — |
| Kagi LLM Benchmark | 53.2% | — |
| DTBench | 82.7% | — |
| LMCA | 24.3% | — |
| Epoch Capabilities Index | 139.92 | — |
| ForecastBench | 58 | — |
Math DeepSeek-V3.1 leads
DeepSeek-V3.1: 38.9 (#122), Tulu 3 (Tülu 3) 70B: 14.2 (#303)
| Benchmark | DeepSeek-V3.1 | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Math | 1420 | 1242 |
| OTIS Mock AIME 2024-2025 | — | 4.4% |
| MATH Level 5 | — | 42.7% |
Knowledge DeepSeek-V3.1 leads
DeepSeek-V3.1: 43.7 (#90), Tulu 3 (Tülu 3) 70B: 25.0 (#264)
| Benchmark | DeepSeek-V3.1 | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| GPQA Diamond | — | 46.3% |
| Vectara Hallucination Rate | 5.5% | — |
| LMArena Expert | 1405 | — |
Multilingual DeepSeek-V3.1 leads
DeepSeek-V3.1: 51.6 (#106), Tulu 3 (Tülu 3) 70B: 39.9 (#222)
| Benchmark | DeepSeek-V3.1 | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Non-English | 1400 | 1236 |
| LMArena Chinese | 1469 | 1249 |
| LMArena Russian | 1405 | 1246 |
| LMArena French | 1447 | — |
| LMArena German | 1411 | — |
| LMArena Japanese | 1378 | — |
| LMArena Korean | 1337 | — |
| LMArena Spanish | 1431 | — |
Instruction Following DeepSeek-V3.1 leads
DeepSeek-V3.1: 73.9 (#110), Tulu 3 (Tülu 3) 70B: 64.8 (#227)
| Benchmark | DeepSeek-V3.1 | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Instruction Following | 1400 | 1233 |
Long Context Too close to call
DeepSeek-V3.1: 36.3 (#232), Tulu 3 (Tülu 3) 70B: 37.1 (#222)
| Benchmark | DeepSeek-V3.1 | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Longer Query | 1422 | 1224 |
| Fiction.LiveBench | 52.8% | — |
Writing & Preference DeepSeek-V3.1 leads
DeepSeek-V3.1: 60.3 (#98), Tulu 3 (Tülu 3) 70B: 45.6 (#223)
| Benchmark | DeepSeek-V3.1 | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Text | 1420 | 1256 |
| LMArena Creative Writing | 1401 | 1231 |
| LMArena Multi-Turn | 1408 | 1252 |
| EQ-Bench Creative Writing | 1436 | — |
Frequently asked questions
Is DeepSeek-V3.1 better than Tulu 3 (Tülu 3) 70B?
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 33.0 on the Noometry Index.
Is DeepSeek-V3.1 or Tulu 3 (Tülu 3) 70B better for coding?
DeepSeek-V3.1 scores higher on coding benchmarks: 40.3 versus 36.0 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.1 and Tulu 3 (Tülu 3) 70B share?
11 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Tulu 3 (Tülu 3) 70B has 14.