Model comparison
Dolly 2.0-12b vs Llama 13b
Dolly 2.0-12b is the stronger model overall, scoring 25.5 to 24.4 on the Noometry Index.
Last verified . 16 shared benchmarks.
Databricks
MetaSummary
- They share 16 benchmarks with published results for both. Dolly 2.0-12b scores higher in 6 categories and Llama 13b in 0 categories; 4 gaps are clear of the uncertainty.
Side by side
| Dolly 2.0-12b | Llama 13b | |
|---|---|---|
| Provider | Databricks | Meta |
| Noometry Index | 25.5 | 24.4 |
| Released | 2023-04-11 | 2023-02-24 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 17 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Dolly 2.0-12b leads
Dolly 2.0-12b: 23.4 (#332), Llama 13b: 21.4 (#337)
| Benchmark | Dolly 2.0-12b | Llama 13b |
|---|---|---|
| LMArena Coding | 776 | 683 |
Reasoning Dolly 2.0-12b leads
Dolly 2.0-12b: 15.3 (#316), Llama 13b: 14.0 (#329)
| Benchmark | Dolly 2.0-12b | Llama 13b |
|---|---|---|
| LMArena Hard Prompts | 804 | 728 |
| Epoch Capabilities Index | 89.67 | 100.58 |
| HellaSwag | 70.8% | 79.2% |
| PIQA | 75.4% | 80.1% |
| WinoGrande | 61.8% | 73% |
| BIG-Bench Hard | — | 37.9% |
| LAMBADA | — | 75.2% |
Math Too close to call
Dolly 2.0-12b: 27.3 (#251), Llama 13b: 26.7 (#256)
| Benchmark | Dolly 2.0-12b | Llama 13b |
|---|---|---|
| LMArena Math | 871 | 838 |
| GSM8K | — | 20.6% |
Knowledge Not comparable
Dolly 2.0-12b: —, Llama 13b: —
| Benchmark | Dolly 2.0-12b | Llama 13b |
|---|---|---|
| ARC (AI2) Challenge | 39.6% | 52.7% |
| BoolQ | 56.3% | 78.7% |
| MMLU | 26.2% | 47.7% |
| OpenBookQA | 39.2% | 56.4% |
| TriviaQA | — | 77.9% |
Multimodal Not comparable
Dolly 2.0-12b: —, Llama 13b: —
| Benchmark | Dolly 2.0-12b | Llama 13b |
|---|---|---|
| ScienceQA | — | 43.3% |
Multilingual Too close to call
Dolly 2.0-12b: 17.4 (#296), Llama 13b: 16.6 (#297)
| Benchmark | Dolly 2.0-12b | Llama 13b |
|---|---|---|
| LMArena Non-English | 836 | 819 |
| LMArena Chinese | 836 | — |
Instruction Following Dolly 2.0-12b leads
Dolly 2.0-12b: 38.7 (#304), Llama 13b: 36.7 (#305)
| Benchmark | Dolly 2.0-12b | Llama 13b |
|---|---|---|
| LMArena Instruction Following | 814 | 781 |
Writing & Preference Dolly 2.0-12b leads
Dolly 2.0-12b: 15.2 (#311), Llama 13b: 13.8 (#312)
| Benchmark | Dolly 2.0-12b | Llama 13b |
|---|---|---|
| LMArena Text | 851 | 834 |
| LMArena Creative Writing | 864 | 794 |
| LMArena Multi-Turn | 740 | 753 |
Frequently asked questions
Is Dolly 2.0-12b better than Llama 13b?
Dolly 2.0-12b is the stronger model overall, scoring 25.5 to 24.4 on the Noometry Index.
Is Dolly 2.0-12b or Llama 13b better for coding?
Dolly 2.0-12b scores higher on coding benchmarks: 23.4 versus 21.4 in the Noometry coding category.
How many benchmarks do Dolly 2.0-12b and Llama 13b share?
16 benchmarks have published results for both models. Dolly 2.0-12b has 17 scored results on Noometry and Llama 13b has 21.