Model comparison
DeepSeek LLM 67B vs DeepSeek-V3.2-Speciale
DeepSeek-V3.2-Speciale is the stronger model overall, scoring 39.1 to 24.8 on the Noometry Index.
Last verified . 0 shared benchmarks.
DeepSeekSummary
- The widest gap is in reasoning, where DeepSeek-V3.2-Speciale leads 32.1 to 15.9.
Side by side
| DeepSeek LLM 67B | DeepSeek-V3.2-Speciale | |
|---|---|---|
| Provider | DeepSeek | DeepSeek |
| Noometry Index | 24.8 | 39.1 |
| Released | 2023-11-29 | 2025-12-01 |
| Weights | Open | Open |
| Context window | — | 128K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.58 |
| Output $ / M tokens | — | $1.68 |
| Results tracked | 15 | 3 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.2-Speciale leads
DeepSeek LLM 67B: 31.3 (#273), DeepSeek-V3.2-Speciale: 39.9 (#136)
| Benchmark | DeepSeek LLM 67B | DeepSeek-V3.2-Speciale |
|---|---|---|
| WeirdML | — | 46.7% |
| LMArena Coding | 1096 | — |
Reasoning DeepSeek-V3.2-Speciale leads
DeepSeek LLM 67B: 15.9 (#300), DeepSeek-V3.2-Speciale: 32.1 (#65)
| Benchmark | DeepSeek LLM 67B | DeepSeek-V3.2-Speciale |
|---|---|---|
| SimpleBench | — | 52.6% |
| Chess Puzzles | 0% | — |
| LMArena Hard Prompts | 1070 | — |
| Epoch Capabilities Index | 110.54 | — |
Math Not comparable
DeepSeek LLM 67B: 8.1 (#318), DeepSeek-V3.2-Speciale: —
| Benchmark | DeepSeek LLM 67B | DeepSeek-V3.2-Speciale |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | — |
| LMArena Math | 1108 | — |
| MATH Level 5 | 6.4% | — |
Knowledge Not comparable
DeepSeek LLM 67B: 7.1 (#308), DeepSeek-V3.2-Speciale: —
| Benchmark | DeepSeek LLM 67B | DeepSeek-V3.2-Speciale |
|---|---|---|
| GPQA Diamond | 24.6% | — |
Multilingual Not comparable
DeepSeek LLM 67B: 29.5 (#262), DeepSeek-V3.2-Speciale: —
| Benchmark | DeepSeek LLM 67B | DeepSeek-V3.2-Speciale |
|---|---|---|
| LMArena Non-English | 1073 | — |
| LMArena Chinese | 1132 | — |
Instruction Following Not comparable
DeepSeek LLM 67B: 55.7 (#272), DeepSeek-V3.2-Speciale: —
| Benchmark | DeepSeek LLM 67B | DeepSeek-V3.2-Speciale |
|---|---|---|
| LMArena Instruction Following | 1079 | — |
Long Context Not comparable
DeepSeek LLM 67B: 33.1 (#260), DeepSeek-V3.2-Speciale: —
| Benchmark | DeepSeek LLM 67B | DeepSeek-V3.2-Speciale |
|---|---|---|
| LMArena Longer Query | 1092 | — |
Writing & Preference DeepSeek-V3.2-Speciale leads
DeepSeek LLM 67B: 31.8 (#277), DeepSeek-V3.2-Speciale: 46.1 (#217)
| Benchmark | DeepSeek LLM 67B | DeepSeek-V3.2-Speciale |
|---|---|---|
| LMArena Text | 1105 | — |
| LMArena Creative Writing | 1067 | — |
| EQ-Bench Creative Writing | — | 1276 |
| LMArena Multi-Turn | 1082 | — |
Frequently asked questions
Is DeepSeek LLM 67B better than DeepSeek-V3.2-Speciale?
DeepSeek-V3.2-Speciale is the stronger model overall, scoring 39.1 to 24.8 on the Noometry Index.
Is DeepSeek LLM 67B or DeepSeek-V3.2-Speciale better for coding?
DeepSeek-V3.2-Speciale scores higher on coding benchmarks: 39.9 versus 31.3 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and DeepSeek-V3.2-Speciale share?
0 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and DeepSeek-V3.2-Speciale has 3.