# DeepSeek-V3.1 vs Dolly 2.0-12b

> DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 25.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v3-1-vs-dolly-2-0-12b
- Last updated: 2026-10-12
- Shared benchmarks: 10

## Summary

- They share 10 benchmarks with published results for both. DeepSeek-V3.1 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 writing & preference, where DeepSeek-V3.1 leads 60.3 to 15.2.

## Snapshot

| | DeepSeek-V3.1 | Dolly 2.0-12b |
|---|---|---|
| Provider | DeepSeek | Databricks |
| Noometry Index | 42.8 | 25.5 |
| Rank | 108 | 342 |
| Context | 164K | — |
| Input $/M | $0.25 | — |
| Output $/M | $0.95 | — |
| Weights | Open | Open |

## Coding

- DeepSeek-V3.1: 40.3 (#144)
- Dolly 2.0-12b: 23.4 (#332)

| Benchmark | DeepSeek-V3.1 | Dolly 2.0-12b |
|---|---|---|
| LMArena Coding | 1417 | 776 |
| WeirdML | 38.4% | — |

## Reasoning

- DeepSeek-V3.1: 27.9 (#110)
- Dolly 2.0-12b: 15.3 (#316)

| Benchmark | DeepSeek-V3.1 | Dolly 2.0-12b |
|---|---|---|
| LMArena Hard Prompts | 1417 | 804 |
| Epoch Capabilities Index | 139.92 | 89.67 |
| SimpleBench | 40% | — |
| Kagi LLM Benchmark | 53.2% | — |
| DTBench | 82.7% | — |
| LMCA | 24.3% | — |
| ForecastBench | 58 | — |
| HellaSwag | — | 70.8% |
| PIQA | — | 75.4% |
| WinoGrande | — | 61.8% |

## Math

- DeepSeek-V3.1: 38.9 (#122)
- Dolly 2.0-12b: 27.3 (#251)

| Benchmark | DeepSeek-V3.1 | Dolly 2.0-12b |
|---|---|---|
| LMArena Math | 1420 | 871 |

## Knowledge

- DeepSeek-V3.1: 43.7 (#90)
- Dolly 2.0-12b: —

| Benchmark | DeepSeek-V3.1 | Dolly 2.0-12b |
|---|---|---|
| Vectara Hallucination Rate | 5.5% | — |
| LMArena Expert | 1405 | — |
| ARC (AI2) Challenge | — | 39.6% |
| BoolQ | — | 56.3% |
| MMLU | — | 26.2% |
| OpenBookQA | — | 39.2% |

## Multilingual

- DeepSeek-V3.1: 51.6 (#106)
- Dolly 2.0-12b: 17.4 (#296)

| Benchmark | DeepSeek-V3.1 | Dolly 2.0-12b |
|---|---|---|
| LMArena Non-English | 1400 | 836 |
| LMArena Chinese | 1469 | 836 |
| LMArena French | 1447 | — |
| LMArena German | 1411 | — |
| LMArena Japanese | 1378 | — |
| LMArena Korean | 1337 | — |
| LMArena Russian | 1405 | — |
| LMArena Spanish | 1431 | — |

## Instruction Following

- DeepSeek-V3.1: 73.9 (#110)
- Dolly 2.0-12b: 38.7 (#304)

| Benchmark | DeepSeek-V3.1 | Dolly 2.0-12b |
|---|---|---|
| LMArena Instruction Following | 1400 | 814 |

## Long Context

- DeepSeek-V3.1: 36.3 (#232)
- Dolly 2.0-12b: —

| Benchmark | DeepSeek-V3.1 | Dolly 2.0-12b |
|---|---|---|
| Fiction.LiveBench | 52.8% | — |
| LMArena Longer Query | 1422 | — |

## Writing & Preference

- DeepSeek-V3.1: 60.3 (#98)
- Dolly 2.0-12b: 15.2 (#311)

| Benchmark | DeepSeek-V3.1 | Dolly 2.0-12b |
|---|---|---|
| LMArena Text | 1420 | 851 |
| LMArena Creative Writing | 1401 | 864 |
| LMArena Multi-Turn | 1408 | 740 |
| EQ-Bench Creative Writing | 1436 | — |

## FAQ

### Is DeepSeek-V3.1 better than Dolly 2.0-12b?

DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 25.5 on the Noometry Index.

### Is DeepSeek-V3.1 or Dolly 2.0-12b better for coding?

DeepSeek-V3.1 scores higher on coding benchmarks: 40.3 versus 23.4 in the Noometry coding category.

### How many benchmarks do DeepSeek-V3.1 and Dolly 2.0-12b share?

10 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Dolly 2.0-12b has 17.