Databricks
DBRX benchmark results
7 results on 5 benchmarks from 3 sources.
Open weights Released: Mar 27, 2024 Updated Oct 22, 2026
Also listed as: dbrx-instruct · DBRX Instruct · databricks/dbrx-instruct · databricks/dbrx-base
All benchmark results
| Benchmarks | Score | Rank | Setting | Source type | Date |
|---|---|---|---|---|---|
| Open LLM Leaderboard v2 | 25.2% | 1607 | Average of 6 normalised benchmarks; bfloat16; 131.597B params“databricks/dbrx-instruct” | Official leaderboard | Mar 13, 2025 |
| Open LLM Leaderboard v2 | 16.4% | 3156 | Average of 6 normalised benchmarks; float16“databricks/dbrx-base” | Official leaderboard | Mar 13, 2025 |
| GPQA Diamond | 32.9% | 256 | Epoch AI run, accuracy | Independent run | Jan 27, 2025 |
| MMLU | 74.1% | 42 | Exact match; 5-shot“DBRX Instruct” | Official leaderboard via Epoch AI | — |
| HumanEval / HumanEval+ | 75.0% | 32 | HumanEval (base tests), pass@1 greedy“databricks/dbrx-instruct” | Official leaderboard | Dec 26, 2024 |
| HumanEval / HumanEval+ | 70.1% | 32 | HumanEval+ (extended tests), pass@1 greedy“databricks/dbrx-instruct” | Official leaderboard | Dec 26, 2024 |
| MATH Level 5 | 11.7% | 95 | Epoch AI run, accuracy | Independent run | Jan 27, 2025 |
FAQ
Where do these DBRX scores come from?
From Epoch AI Benchmarking Hub, EvalPlus leaderboard, and Hugging Face Open LLM Leaderboard v2 (archived). Each row links to the original leaderboard, dataset or report, and vendor-reported numbers are labelled.