Model comparison
DBRX vs DeepSeek-V3.2-Speciale
DeepSeek-V3.2-Speciale is the stronger model overall, scoring 39.7 to 29.4 on the Noometry Index.
Last verified . 0 shared benchmarks.
Databricks
DeepSeekSummary
- The widest gap is in writing & preference, where DeepSeek-V3.2-Speciale leads 46.0 to 33.6.
Side by side
| DBRX | DeepSeek-V3.2-Speciale | |
|---|---|---|
| Provider | Databricks | DeepSeek |
| Noometry Index | 29.4 | 39.7 |
| Released | 2024-03-27 | 2025-12-01 |
| Weights | Open | Open |
| Context window | — | 128K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.58 |
| Output $ / M tokens | — | $1.68 |
| Results tracked | 21 | 3 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.2-Speciale leads
DBRX: 32.9 (#266), DeepSeek-V3.2-Speciale: 40.4 (#140)
| Benchmark | DBRX | DeepSeek-V3.2-Speciale |
|---|---|---|
| WeirdML | — | 46.7% |
| LMArena Coding | 1132 | — |
| HumanEval+ | 70.1% | — |
| MBPP+ | 55.8% | — |
Reasoning DeepSeek-V3.2-Speciale leads
DBRX: 21.4 (#222), DeepSeek-V3.2-Speciale: 32.9 (#73)
| Benchmark | DBRX | DeepSeek-V3.2-Speciale |
|---|---|---|
| SimpleBench | — | 52.6% |
| LMArena Hard Prompts | 1113 | — |
Math Not comparable
DBRX: 24.3 (#269), DeepSeek-V3.2-Speciale: —
| Benchmark | DBRX | DeepSeek-V3.2-Speciale |
|---|---|---|
| LMArena Math | 1145 | — |
| MATH Level 5 | 11.7% | — |
Knowledge Not comparable
DBRX: 14.9 (#294), DeepSeek-V3.2-Speciale: —
| Benchmark | DBRX | DeepSeek-V3.2-Speciale |
|---|---|---|
| GPQA Diamond | 32.9% | — |
| LMArena Expert | 1076 | — |
Multilingual Not comparable
DBRX: 29.3 (#268), DeepSeek-V3.2-Speciale: —
| Benchmark | DBRX | DeepSeek-V3.2-Speciale |
|---|---|---|
| LMArena Non-English | 1071 | — |
| LMArena Chinese | 1068 | — |
| LMArena French | 1096 | — |
| LMArena German | 1057 | — |
| LMArena Japanese | 990 | — |
| LMArena Korean | 993 | — |
| LMArena Russian | 1078 | — |
| LMArena Spanish | 1064 | — |
Instruction Following Not comparable
DBRX: 57.5 (#270), DeepSeek-V3.2-Speciale: —
| Benchmark | DBRX | DeepSeek-V3.2-Speciale |
|---|---|---|
| LMArena Instruction Following | 1112 | — |
Long Context Not comparable
DBRX: 33.7 (#258), DeepSeek-V3.2-Speciale: —
| Benchmark | DBRX | DeepSeek-V3.2-Speciale |
|---|---|---|
| LMArena Longer Query | 1112 | — |
Writing & Preference DeepSeek-V3.2-Speciale leads
DBRX: 33.6 (#275), DeepSeek-V3.2-Speciale: 46.0 (#222)
| Benchmark | DBRX | DeepSeek-V3.2-Speciale |
|---|---|---|
| LMArena Text | 1119 | — |
| LMArena Creative Writing | 1104 | — |
| EQ-Bench Creative Writing | — | 1276 |
| LMArena Multi-Turn | 1111 | — |
Frequently asked questions
Is DBRX better than DeepSeek-V3.2-Speciale?
DeepSeek-V3.2-Speciale is the stronger model overall, scoring 39.7 to 29.4 on the Noometry Index.
Is DBRX or DeepSeek-V3.2-Speciale better for coding?
DeepSeek-V3.2-Speciale scores higher on coding benchmarks: 40.4 versus 32.9 in the Noometry coding category.
How many benchmarks do DBRX and DeepSeek-V3.2-Speciale share?
0 benchmarks have published results for both models. DBRX has 21 scored results on Noometry and DeepSeek-V3.2-Speciale has 3.