# DBRX vs DeepSeek-V3

> DeepSeek-V3 is the stronger model overall, scoring 39.5 to 29.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/dbrx-vs-deepseek-v3
- Last updated: 2026-10-11
- Shared benchmarks: 21

## Summary

- They share 21 benchmarks with published results for both. DBRX scores higher in 1 category and DeepSeek-V3 in 7 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V3 leads 57.4 to 33.6.
- The biggest single-benchmark swing is MATH Level 5: 11.7% for DBRX and 75.5% for DeepSeek-V3.

## Snapshot

| | DBRX | DeepSeek-V3 |
|---|---|---|
| Provider | Databricks | DeepSeek |
| Noometry Index | 29.4 | 39.5 |
| Rank | 311 | 166 |
| Context | — | 164K |
| Input $/M | — | $0.24 |
| Output $/M | — | $0.90 |
| Weights | Open | Open |

## Coding

- DBRX: 32.9 (#266)
- DeepSeek-V3: 42.3 (#106)

| Benchmark | DBRX | DeepSeek-V3 |
|---|---|---|
| LMArena Coding | 1132 | 1368 |
| HumanEval+ | 70.1% | 86.6% |
| MBPP+ | 55.8% | 73% |
| Aider Polyglot | — | 55.1% |
| SciCode | — | 35.8% |
| WeirdML | — | 36.1% |
| BigCodeBench Instruct | — | 50% |
| LiveBench Coding | — | 70.9% |
| BigCodeBench Complete | — | 62.2% |

## Agentic & Tool Use

- DBRX: —
- DeepSeek-V3: —

| Benchmark | DBRX | DeepSeek-V3 |
|---|---|---|
| METR Time Horizons | — | 49.6% |

## Reasoning

- DBRX: 21.4 (#222)
- DeepSeek-V3: 20.5 (#236)

| Benchmark | DBRX | DeepSeek-V3 |
|---|---|---|
| LMArena Hard Prompts | 1113 | 1365 |
| SimpleBench | — | 27.2% |
| Kagi LLM Benchmark | — | 52.3% |
| CritPt | — | 0% |
| LiveBench Reasoning | — | 65.8% |
| DTBench | — | 64.8% |
| LiveBench Data Analysis | — | 60.9% |
| LMCA | — | 15.5% |
| BIG-Bench Hard | — | 87.5% |
| Epoch Capabilities Index | — | 135.94 |
| ForecastBench | — | 59.1 |
| HellaSwag | — | 88.9% |
| LiveBench | — | 66.9% |
| PIQA | — | 84.7% |
| WinoGrande | — | 85.2% |

## Math

- DBRX: 24.3 (#269)
- DeepSeek-V3: 32.1 (#219)

| Benchmark | DBRX | DeepSeek-V3 |
|---|---|---|
| LMArena Math | 1145 | 1373 |
| MATH Level 5 | 11.7% | 75.5% |
| OTIS Mock AIME 2024-2025 | — | 37.8% |
| Omni-MATH | — | 40.3% |
| LiveBench Math | — | 73.5% |
| FrontierMath (Feb 2025 set) | — | 1.7% |

## Knowledge

- DBRX: 14.9 (#294)
- DeepSeek-V3: 37.5 (#155)

| Benchmark | DBRX | DeepSeek-V3 |
|---|---|---|
| GPQA Diamond | 32.9% | 67.6% |
| LMArena Expert | 1076 | 1351 |
| MMLU-Pro | — | 72.3% |
| Confabulations | — | 26.1% |
| Vectara Hallucination Rate | — | 6.1% |
| GPQA (HELM) | — | 53.8% |
| ARC (AI2) Challenge | — | 95.3% |
| MMLU | — | 87.2% |
| TriviaQA | — | 82.9% |

## Multilingual

- DBRX: 29.3 (#268)
- DeepSeek-V3: 48.5 (#143)

| Benchmark | DBRX | DeepSeek-V3 |
|---|---|---|
| LMArena Non-English | 1071 | 1358 |
| LMArena Chinese | 1068 | 1391 |
| LMArena French | 1096 | 1385 |
| LMArena German | 1057 | 1374 |
| LMArena Japanese | 990 | 1333 |
| LMArena Korean | 993 | 1319 |
| LMArena Russian | 1078 | 1373 |
| LMArena Spanish | 1064 | 1358 |

## Instruction Following

- DBRX: 57.5 (#270)
- DeepSeek-V3: 72.8 (#130)

| Benchmark | DBRX | DeepSeek-V3 |
|---|---|---|
| LMArena Instruction Following | 1112 | 1345 |
| LiveBench Instruction Following | — | 81.5% |
| IFEval | — | 83.2% |

## Long Context

- DBRX: 33.7 (#258)
- DeepSeek-V3: 34.0 (#253)

| Benchmark | DBRX | DeepSeek-V3 |
|---|---|---|
| LMArena Longer Query | 1112 | 1352 |
| Fiction.LiveBench | — | 50% |

## Writing & Preference

- DBRX: 33.6 (#275)
- DeepSeek-V3: 57.4 (#130)

| Benchmark | DBRX | DeepSeek-V3 |
|---|---|---|
| LMArena Text | 1119 | 1375 |
| LMArena Creative Writing | 1104 | 1364 |
| LMArena Multi-Turn | 1111 | 1389 |
| Short-Story Creative Writing | — | 77% |
| EQ-Bench Creative Writing | — | 1472 |
| WildBench | — | 83% |
| LiveBench Language | — | 49.1% |

## FAQ

### Is DBRX better than DeepSeek-V3?

DeepSeek-V3 is the stronger model overall, scoring 39.5 to 29.4 on the Noometry Index.

### Is DBRX or DeepSeek-V3 better for coding?

DeepSeek-V3 scores higher on coding benchmarks: 42.3 versus 32.9 in the Noometry coding category.

### How many benchmarks do DBRX and DeepSeek-V3 share?

21 benchmarks have published results for both models. DBRX has 21 scored results on Noometry and DeepSeek-V3 has 60.