Model comparison
DBRX vs StarCoder 2 15B
DBRX has enough public results to be ranked (#307); StarCoder 2 15B does not yet, so treat this comparison as directional.
Last verified . 1 shared benchmarks.
Databricks
NVIDIASummary
- They share 1 benchmark with published results for both. DBRX scores higher in 0 categories and StarCoder 2 15B in 1 category; one gap is clear of the uncertainty.
- The widest gap is in coding, where StarCoder 2 15B leads 36.0 to 32.3.
Side by side
| DBRX | StarCoder 2 15B | |
|---|---|---|
| Provider | Databricks | NVIDIA |
| Noometry Index | 29.0 | 37.5 |
| Released | 2024-03-27 | 2024-02-20 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 21 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding StarCoder 2 15B leads
DBRX: 32.3 (#262), StarCoder 2 15B: 36.0
| Benchmark | DBRX | StarCoder 2 15B |
|---|---|---|
| HumanEval+ | 70.1% | 37.8% |
| LMArena Coding | 1132 | — |
| BigCodeBench Complete | — | 38.4% |
| MBPP+ | 55.8% | — |
Reasoning Not comparable
DBRX: 20.5 (#218), StarCoder 2 15B: —
| Benchmark | DBRX | StarCoder 2 15B |
|---|---|---|
| LMArena Hard Prompts | 1113 | — |
| Epoch Capabilities Index | — | 104.61 |
| WinoGrande | — | 64.3% |
Math Not comparable
DBRX: 22.3 (#265), StarCoder 2 15B: —
| Benchmark | DBRX | StarCoder 2 15B |
|---|---|---|
| LMArena Math | 1145 | — |
| MATH Level 5 | 11.7% | — |
| GSM8K | — | 57.7% |
Knowledge Not comparable
DBRX: 14.9 (#289), StarCoder 2 15B: —
| Benchmark | DBRX | StarCoder 2 15B |
|---|---|---|
| GPQA Diamond | 32.9% | — |
| LMArena Expert | 1076 | — |
| ARC (AI2) Challenge | — | 47.2% |
| MMLU | — | 64.1% |
Multilingual Not comparable
DBRX: 29.4 (#263), StarCoder 2 15B: —
| Benchmark | DBRX | StarCoder 2 15B |
|---|---|---|
| LMArena Non-English | 1071 | — |
| LMArena Chinese | 1068 | — |
| LMArena French | 1095 | — |
| LMArena German | 1057 | — |
| LMArena Japanese | 990 | — |
| LMArena Korean | 993 | — |
| LMArena Russian | 1078 | — |
| LMArena Spanish | 1064 | — |
Instruction Following Not comparable
DBRX: 57.7 (#265), StarCoder 2 15B: —
| Benchmark | DBRX | StarCoder 2 15B |
|---|---|---|
| LMArena Instruction Following | 1112 | — |
Long Context Not comparable
DBRX: 33.7 (#253), StarCoder 2 15B: —
| Benchmark | DBRX | StarCoder 2 15B |
|---|---|---|
| LMArena Longer Query | 1112 | — |
Writing & Preference Not comparable
DBRX: 33.7 (#271), StarCoder 2 15B: —
| Benchmark | DBRX | StarCoder 2 15B |
|---|---|---|
| LMArena Text | 1119 | — |
| LMArena Creative Writing | 1104 | — |
| LMArena Multi-Turn | 1111 | — |
Frequently asked questions
Is DBRX better than StarCoder 2 15B?
DBRX has enough public results to be ranked (#307); StarCoder 2 15B does not yet, so treat this comparison as directional.
Is DBRX or StarCoder 2 15B better for coding?
StarCoder 2 15B scores higher on coding benchmarks: 36.0 versus 32.3 in the Noometry coding category.
How many benchmarks do DBRX and StarCoder 2 15B share?
1 benchmark has published results for both models. DBRX has 21 scored results on Noometry and StarCoder 2 15B has 7.