Model comparison
DBRX vs Qwen1.5-32B
Qwen1.5-32B is the stronger model overall, scoring 30.0 to 29.0 on the Noometry Index.
Last verified . 18 shared benchmarks.
Databricks
Alibaba (Qwen)Summary
- They share 18 benchmarks with published results for both. DBRX scores higher in 2 categories and Qwen1.5-32B in 6 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen1.5-32B leads 30.5 to 22.3.
Side by side
| DBRX | Qwen1.5-32B | |
|---|---|---|
| Provider | Databricks | Alibaba (Qwen) |
| Noometry Index | 29.0 | 30.0 |
| Released | 2024-03-27 | 2024-02-04 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 21 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DBRX leads
DBRX: 32.3 (#262), Qwen1.5-32B: 31.1 (#278)
| Benchmark | DBRX | Qwen1.5-32B |
|---|---|---|
| LMArena Coding | 1132 | 1155 |
| BigCodeBench Instruct | — | 32.3% |
| BigCodeBench Complete | — | 42% |
| HumanEval+ | 70.1% | — |
| MBPP+ | 55.8% | — |
Reasoning Too close to call
DBRX: 20.5 (#218), Qwen1.5-32B: 20.9 (#210)
| Benchmark | DBRX | Qwen1.5-32B |
|---|---|---|
| LMArena Hard Prompts | 1113 | 1131 |
Math Qwen1.5-32B leads
DBRX: 22.3 (#265), Qwen1.5-32B: 30.5 (#207)
| Benchmark | DBRX | Qwen1.5-32B |
|---|---|---|
| LMArena Math | 1145 | 1155 |
| MATH Level 5 | 11.7% | — |
Knowledge DBRX leads
DBRX: 14.9 (#289), Qwen1.5-32B: 13.5 (#291)
| Benchmark | DBRX | Qwen1.5-32B |
|---|---|---|
| GPQA Diamond | 32.9% | 30.7% |
| LMArena Expert | 1076 | 1126 |
| MMLU | — | 74.4% |
Multilingual Qwen1.5-32B leads
DBRX: 29.4 (#263), Qwen1.5-32B: 31.6 (#254)
| Benchmark | DBRX | Qwen1.5-32B |
|---|---|---|
| LMArena Non-English | 1071 | 1106 |
| LMArena Chinese | 1068 | 1176 |
| LMArena French | 1095 | 1100 |
| LMArena German | 1057 | 1058 |
| LMArena Japanese | 990 | 1027 |
| LMArena Korean | 993 | 1007 |
| LMArena Russian | 1078 | 1073 |
| LMArena Spanish | 1064 | 1089 |
Instruction Following Too close to call
DBRX: 57.7 (#265), Qwen1.5-32B: 58.0 (#260)
| Benchmark | DBRX | Qwen1.5-32B |
|---|---|---|
| LMArena Instruction Following | 1112 | 1116 |
Long Context Qwen1.5-32B leads
DBRX: 33.7 (#253), Qwen1.5-32B: 34.7 (#241)
| Benchmark | DBRX | Qwen1.5-32B |
|---|---|---|
| LMArena Longer Query | 1112 | 1146 |
Writing & Preference Too close to call
DBRX: 33.7 (#271), Qwen1.5-32B: 34.4 (#266)
| Benchmark | DBRX | Qwen1.5-32B |
|---|---|---|
| LMArena Text | 1119 | 1137 |
| LMArena Creative Writing | 1104 | 1083 |
| LMArena Multi-Turn | 1111 | 1140 |
Frequently asked questions
Is DBRX better than Qwen1.5-32B?
Qwen1.5-32B is the stronger model overall, scoring 30.0 to 29.0 on the Noometry Index.
Is DBRX or Qwen1.5-32B better for coding?
DBRX scores higher on coding benchmarks: 32.3 versus 31.1 in the Noometry coding category.
How many benchmarks do DBRX and Qwen1.5-32B share?
18 benchmarks have published results for both models. DBRX has 21 scored results on Noometry and Qwen1.5-32B has 21.