Model comparison
Deepseek Coder v2 vs Dolly 2.0-12b
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 25.5 on the Noometry Index.
Last verified . 11 shared benchmarks.
DeepSeek
DatabricksSummary
- They share 11 benchmarks with published results for both. Deepseek Coder v2 scores higher in 6 categories and Dolly 2.0-12b in 0 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where Deepseek Coder v2 leads 61.7 to 38.7.
Side by side
| Deepseek Coder v2 | Dolly 2.0-12b | |
|---|---|---|
| Provider | DeepSeek | Databricks |
| Noometry Index | 35.9 | 25.5 |
| Released | 2024-06-17 | 2023-04-11 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 24 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Deepseek Coder v2 leads
Deepseek Coder v2: 38.1 (#183), Dolly 2.0-12b: 23.4 (#332)
| Benchmark | Deepseek Coder v2 | Dolly 2.0-12b |
|---|---|---|
| LMArena Coding | 1251 | 776 |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 59.7% | — |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |
Reasoning Deepseek Coder v2 leads
Deepseek Coder v2: 23.6 (#176), Dolly 2.0-12b: 15.3 (#316)
| Benchmark | Deepseek Coder v2 | Dolly 2.0-12b |
|---|---|---|
| LMArena Hard Prompts | 1207 | 804 |
| WinoGrande | 83.7% | 61.8% |
| Epoch Capabilities Index | — | 89.67 |
| HellaSwag | — | 70.8% |
| PIQA | — | 75.4% |
Math Deepseek Coder v2 leads
Deepseek Coder v2: 34.9 (#190), Dolly 2.0-12b: 27.3 (#251)
| Benchmark | Deepseek Coder v2 | Dolly 2.0-12b |
|---|---|---|
| LMArena Math | 1241 | 871 |
| GSM8K | 94.5% | — |
Knowledge Not comparable
Deepseek Coder v2: 32.3 (#212), Dolly 2.0-12b: —
| Benchmark | Deepseek Coder v2 | Dolly 2.0-12b |
|---|---|---|
| ARC (AI2) Challenge | 64.3% | 39.6% |
| LMArena Expert | 1181 | — |
| BoolQ | — | 56.3% |
| MMLU | — | 26.2% |
| OpenBookQA | — | 39.2% |
Multilingual Deepseek Coder v2 leads
Deepseek Coder v2: 36.3 (#240), Dolly 2.0-12b: 17.4 (#296)
| Benchmark | Deepseek Coder v2 | Dolly 2.0-12b |
|---|---|---|
| LMArena Non-English | 1182 | 836 |
| LMArena Chinese | 1201 | 836 |
| LMArena French | 1185 | — |
| LMArena German | 1164 | — |
| LMArena Japanese | 1126 | — |
| LMArena Korean | 1104 | — |
| LMArena Russian | 1188 | — |
| LMArena Spanish | 1153 | — |
Instruction Following Deepseek Coder v2 leads
Deepseek Coder v2: 61.7 (#242), Dolly 2.0-12b: 38.7 (#304)
| Benchmark | Deepseek Coder v2 | Dolly 2.0-12b |
|---|---|---|
| LMArena Instruction Following | 1180 | 814 |
Long Context Not comparable
Deepseek Coder v2: 37.0 (#224), Dolly 2.0-12b: —
| Benchmark | Deepseek Coder v2 | Dolly 2.0-12b |
|---|---|---|
| LMArena Longer Query | 1219 | — |
Writing & Preference Deepseek Coder v2 leads
Deepseek Coder v2: 38.2 (#253), Dolly 2.0-12b: 15.2 (#311)
| Benchmark | Deepseek Coder v2 | Dolly 2.0-12b |
|---|---|---|
| LMArena Text | 1191 | 851 |
| LMArena Creative Writing | 1120 | 864 |
| LMArena Multi-Turn | 1177 | 740 |
Frequently asked questions
Is Deepseek Coder v2 better than Dolly 2.0-12b?
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 25.5 on the Noometry Index.
Is Deepseek Coder v2 or Dolly 2.0-12b better for coding?
Deepseek Coder v2 scores higher on coding benchmarks: 38.1 versus 23.4 in the Noometry coding category.
How many benchmarks do Deepseek Coder v2 and Dolly 2.0-12b share?
11 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Dolly 2.0-12b has 17.