Model comparison
DBRX vs Qwen2.5-Coder-32B
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 29.4 on the Noometry Index.
Last verified . 14 shared benchmarks.
Databricks
Alibaba (Qwen)Summary
- They share 14 benchmarks with published results for both. DBRX scores higher in 2 categories and Qwen2.5-Coder-32B in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen2.5-Coder-32B leads 33.4 to 14.9.
Side by side
| DBRX | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Databricks | Alibaba (Qwen) |
| Noometry Index | 29.4 | 33.4 |
| Released | 2024-03-27 | 2024-09-18 |
| Weights | Open | Open |
| Context window | — | 33K |
| Max output | — | 29K |
| Input $ / M tokens | — | $0.66 |
| Output $ / M tokens | — | $1 |
| Results tracked | 21 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DBRX leads
DBRX: 32.9 (#266), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | DBRX | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Coding | 1132 | 1276 |
| HumanEval+ | 70.1% | 87.2% |
| MBPP+ | 55.8% | 77% |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| BigCodeBench Complete | — | 58% |
Reasoning Too close to call
DBRX: 21.4 (#222), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | DBRX | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1113 | 1251 |
| LiveBench Reasoning | — | 42.1% |
| LiveBench Data Analysis | — | 49.9% |
| Epoch Capabilities Index | — | 119.49 |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |
Math Qwen2.5-Coder-32B leads
DBRX: 24.3 (#269), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | DBRX | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1145 | 1251 |
| LiveBench Math | — | 46.6% |
| MATH Level 5 | 11.7% | — |
| GSM8K | — | 93% |
Knowledge Qwen2.5-Coder-32B leads
DBRX: 14.9 (#294), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | DBRX | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1076 | 1221 |
| GPQA Diamond | 32.9% | — |
| ARC (AI2) Challenge | — | 70.5% |
| MMLU | — | 79.1% |
Multilingual Qwen2.5-Coder-32B leads
DBRX: 29.3 (#268), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | DBRX | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1071 | 1205 |
| LMArena Chinese | 1068 | 1222 |
| LMArena Russian | 1078 | 1228 |
| LMArena French | 1096 | — |
| LMArena German | 1057 | — |
| LMArena Japanese | 990 | — |
| LMArena Korean | 993 | — |
| LMArena Spanish | 1064 | — |
Instruction Following Qwen2.5-Coder-32B leads
DBRX: 57.5 (#270), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | DBRX | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1112 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
Long Context Qwen2.5-Coder-32B leads
DBRX: 33.7 (#258), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | DBRX | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1112 | 1251 |
Writing & Preference Qwen2.5-Coder-32B leads
DBRX: 33.6 (#275), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | DBRX | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1119 | 1230 |
| LMArena Creative Writing | 1104 | 1174 |
| LMArena Multi-Turn | 1111 | 1222 |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is DBRX better than Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 29.4 on the Noometry Index.
Is DBRX or Qwen2.5-Coder-32B better for coding?
DBRX scores higher on coding benchmarks: 32.9 versus 22.6 in the Noometry coding category.
How many benchmarks do DBRX and Qwen2.5-Coder-32B share?
14 benchmarks have published results for both models. DBRX has 21 scored results on Noometry and Qwen2.5-Coder-32B has 31.