Model comparison
DBRX vs Gemma 3 12B
Gemma 3 12B is the stronger model overall, scoring 31.8 to 29.0 on the Noometry Index.
Last verified . 13 shared benchmarks.
Databricks
GoogleSummary
- They share 13 benchmarks with published results for both. DBRX scores higher in 3 categories and Gemma 3 12B in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where Gemma 3 12B leads 45.9 to 29.4.
- The biggest single-benchmark swing is GPQA Diamond: 32.9% for DBRX and 39.5% for Gemma 3 12B.
Side by side
| DBRX | Gemma 3 12B | |
|---|---|---|
| Provider | Databricks | |
| Noometry Index | 29.0 | 31.8 |
| Released | 2024-03-27 | 2025-03-12 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 8K |
| Input $ / M tokens | — | $0.05 |
| Output $ / M tokens | — | $0.15 |
| Results tracked | 21 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DBRX leads
DBRX: 32.3 (#262), Gemma 3 12B: 31.2 (#276)
| Benchmark | DBRX | Gemma 3 12B |
|---|---|---|
| LMArena Coding | 1132 | 1281 |
| SciCode | — | 17.4% |
| HumanEval+ | 70.1% | — |
| MBPP+ | 55.8% | — |
Agentic & Tool Use Not comparable
DBRX: —, Gemma 3 12B: 25.3 (#101)
| Benchmark | DBRX | Gemma 3 12B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 30.4% |
Reasoning DBRX leads
DBRX: 20.5 (#218), Gemma 3 12B: 15.2 (#307)
| Benchmark | DBRX | Gemma 3 12B |
|---|---|---|
| LMArena Hard Prompts | 1113 | 1309 |
| CritPt | — | 0% |
| Chess Puzzles | — | 0% |
| DTBench | — | 48.8% |
| LMCA | — | 4.5% |
| Epoch Capabilities Index | — | 123.49 |
Math DBRX leads
DBRX: 22.3 (#265), Gemma 3 12B: 21.0 (#273)
| Benchmark | DBRX | Gemma 3 12B |
|---|---|---|
| LMArena Math | 1145 | 1307 |
| OTIS Mock AIME 2024-2025 | — | 16.7% |
| MATH Level 5 | 11.7% | — |
Knowledge Gemma 3 12B leads
DBRX: 14.9 (#289), Gemma 3 12B: 26.5 (#250)
| Benchmark | DBRX | Gemma 3 12B |
|---|---|---|
| GPQA Diamond | 32.9% | 39.5% |
| LMArena Expert | 1076 | 1248 |
| Vectara Hallucination Rate | — | 4.4% |
Multimodal Not comparable
DBRX: —, Gemma 3 12B: —
| Benchmark | DBRX | Gemma 3 12B |
|---|---|---|
| MindCube | — | 46.7% |
Multilingual Gemma 3 12B leads
DBRX: 29.4 (#263), Gemma 3 12B: 45.9 (#160)
| Benchmark | DBRX | Gemma 3 12B |
|---|---|---|
| LMArena Non-English | 1071 | 1318 |
| LMArena German | 1057 | 1370 |
| LMArena Russian | 1078 | 1335 |
| LMArena Chinese | 1068 | — |
| LMArena French | 1095 | — |
| LMArena Japanese | 990 | — |
| LMArena Korean | 993 | — |
| LMArena Spanish | 1064 | — |
Instruction Following Gemma 3 12B leads
DBRX: 57.7 (#265), Gemma 3 12B: 68.8 (#179)
| Benchmark | DBRX | Gemma 3 12B |
|---|---|---|
| LMArena Instruction Following | 1112 | 1299 |
Long Context Gemma 3 12B leads
DBRX: 33.7 (#253), Gemma 3 12B: 40.0 (#157)
| Benchmark | DBRX | Gemma 3 12B |
|---|---|---|
| LMArena Longer Query | 1112 | 1317 |
Writing & Preference Gemma 3 12B leads
DBRX: 33.7 (#271), Gemma 3 12B: 47.7 (#205)
| Benchmark | DBRX | Gemma 3 12B |
|---|---|---|
| LMArena Text | 1119 | 1334 |
| LMArena Creative Writing | 1104 | 1331 |
| LMArena Multi-Turn | 1111 | 1334 |
| EQ-Bench Creative Writing | — | 1126 |
Frequently asked questions
Is DBRX better than Gemma 3 12B?
Gemma 3 12B is the stronger model overall, scoring 31.8 to 29.0 on the Noometry Index.
Is DBRX or Gemma 3 12B better for coding?
DBRX scores higher on coding benchmarks: 32.3 versus 31.2 in the Noometry coding category.
How many benchmarks do DBRX and Gemma 3 12B share?
13 benchmarks have published results for both models. DBRX has 21 scored results on Noometry and Gemma 3 12B has 24.