Model comparison
DeepSeek-V3.2-Speciale vs Tulu 3 (Tülu 3) 70B
DeepSeek-V3.2-Speciale is the stronger model overall, scoring 39.7 to 33.0 on the Noometry Index.
Last verified . 0 shared benchmarks.
DeepSeek
Allen Institute for AI (Ai2)Summary
- The widest gap is in reasoning, where DeepSeek-V3.2-Speciale leads 32.9 to 23.9.
Side by side
| DeepSeek-V3.2-Speciale | Tulu 3 (Tülu 3) 70B | |
|---|---|---|
| Provider | DeepSeek | Allen Institute for AI (Ai2) |
| Noometry Index | 39.7 | 33.0 |
| Released | 2025-12-01 | 2024-11-21 |
| Weights | Open | Open |
| Context window | 128K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.58 | — |
| Output $ / M tokens | $1.68 | — |
| Results tracked | 3 | 14 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.2-Speciale leads
DeepSeek-V3.2-Speciale: 40.4 (#140), Tulu 3 (Tülu 3) 70B: 36.0 (#214)
| Benchmark | DeepSeek-V3.2-Speciale | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| WeirdML | 46.7% | — |
| LMArena Coding | — | 1235 |
Reasoning DeepSeek-V3.2-Speciale leads
DeepSeek-V3.2-Speciale: 32.9 (#73), Tulu 3 (Tülu 3) 70B: 23.9 (#169)
| Benchmark | DeepSeek-V3.2-Speciale | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| SimpleBench | 52.6% | — |
| LMArena Hard Prompts | — | 1220 |
Math Not comparable
DeepSeek-V3.2-Speciale: —, Tulu 3 (Tülu 3) 70B: 14.2 (#303)
| Benchmark | DeepSeek-V3.2-Speciale | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 4.4% |
| LMArena Math | — | 1242 |
| MATH Level 5 | — | 42.7% |
Knowledge Not comparable
DeepSeek-V3.2-Speciale: —, Tulu 3 (Tülu 3) 70B: 25.0 (#264)
| Benchmark | DeepSeek-V3.2-Speciale | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| GPQA Diamond | — | 46.3% |
Multilingual Not comparable
DeepSeek-V3.2-Speciale: —, Tulu 3 (Tülu 3) 70B: 39.9 (#222)
| Benchmark | DeepSeek-V3.2-Speciale | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Non-English | — | 1236 |
| LMArena Chinese | — | 1249 |
| LMArena Russian | — | 1246 |
Instruction Following Not comparable
DeepSeek-V3.2-Speciale: —, Tulu 3 (Tülu 3) 70B: 64.8 (#227)
| Benchmark | DeepSeek-V3.2-Speciale | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Instruction Following | — | 1233 |
Long Context Not comparable
DeepSeek-V3.2-Speciale: —, Tulu 3 (Tülu 3) 70B: 37.1 (#222)
| Benchmark | DeepSeek-V3.2-Speciale | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Longer Query | — | 1224 |
Writing & Preference Too close to call
DeepSeek-V3.2-Speciale: 46.0 (#222), Tulu 3 (Tülu 3) 70B: 45.6 (#223)
| Benchmark | DeepSeek-V3.2-Speciale | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Text | — | 1256 |
| LMArena Creative Writing | — | 1231 |
| EQ-Bench Creative Writing | 1276 | — |
| LMArena Multi-Turn | — | 1252 |
Frequently asked questions
Is DeepSeek-V3.2-Speciale better than Tulu 3 (Tülu 3) 70B?
DeepSeek-V3.2-Speciale is the stronger model overall, scoring 39.7 to 33.0 on the Noometry Index.
Is DeepSeek-V3.2-Speciale or Tulu 3 (Tülu 3) 70B better for coding?
DeepSeek-V3.2-Speciale scores higher on coding benchmarks: 40.4 versus 36.0 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.2-Speciale and Tulu 3 (Tülu 3) 70B share?
0 benchmarks have published results for both models. DeepSeek-V3.2-Speciale has 3 scored results on Noometry and Tulu 3 (Tülu 3) 70B has 14.