Model comparison
DBRX vs Mixtral 8x7B
DBRX is the stronger model overall, scoring 29.4 to 27.1 on the Noometry Index.
Last verified . 21 shared benchmarks.
Databricks
Mistral AISummary
- They share 21 benchmarks with published results for both. DBRX scores higher in 6 categories and Mixtral 8x7B in 2 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where DBRX leads 57.5 to 51.0.
Side by side
| DBRX | Mixtral 8x7B | |
|---|---|---|
| Provider | Databricks | Mistral AI |
| Noometry Index | 29.4 | 27.1 |
| Released | 2024-03-27 | 2023-12-11 |
| Weights | Open | Open |
| Context window | — | 32K |
| Max output | — | 32K |
| Input $ / M tokens | — | $0.70 |
| Output $ / M tokens | — | $0.70 |
| Results tracked | 21 | 38 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
DBRX: 32.9 (#266), Mixtral 8x7B: 32.8 (#269)
| Benchmark | DBRX | Mixtral 8x7B |
|---|---|---|
| LMArena Coding | 1132 | 1126 |
| HumanEval+ | 70.1% | 39.6% |
| MBPP+ | 55.8% | 49.7% |
Reasoning DBRX leads
DBRX: 21.4 (#222), Mixtral 8x7B: 18.2 (#285)
| Benchmark | DBRX | Mixtral 8x7B |
|---|---|---|
| LMArena Hard Prompts | 1113 | 1115 |
| DTBench | — | 49.6% |
| Adversarial NLI | — | 55.2% |
| Epoch Capabilities Index | — | 118.47 |
| ForecastBench | — | 56.3 |
| HellaSwag | — | 86.7% |
| PIQA | — | 83.6% |
| WinoGrande | — | 77.2% |
Math DBRX leads
DBRX: 24.3 (#269), Mixtral 8x7B: 18.8 (#289)
| Benchmark | DBRX | Mixtral 8x7B |
|---|---|---|
| LMArena Math | 1145 | 1147 |
| MATH Level 5 | 11.7% | 10% |
| Omni-MATH | — | 10.5% |
| GSM8K | — | 74.4% |
Knowledge DBRX leads
DBRX: 14.9 (#294), Mixtral 8x7B: 11.0 (#301)
| Benchmark | DBRX | Mixtral 8x7B |
|---|---|---|
| GPQA Diamond | 32.9% | 30.6% |
| LMArena Expert | 1076 | 1088 |
| MMLU-Pro | — | 33.5% |
| GPQA (HELM) | — | 29.6% |
| ARC (AI2) Challenge | — | 87.3% |
| MMLU | — | 70.6% |
| OpenBookQA | — | 85.8% |
| TriviaQA | — | 82.2% |
Multilingual Too close to call
DBRX: 29.3 (#268), Mixtral 8x7B: 29.6 (#266)
| Benchmark | DBRX | Mixtral 8x7B |
|---|---|---|
| LMArena Non-English | 1071 | 1077 |
| LMArena Chinese | 1068 | 1055 |
| LMArena French | 1096 | 1166 |
| LMArena German | 1057 | 1114 |
| LMArena Japanese | 990 | 931 |
| LMArena Korean | 993 | 968 |
| LMArena Russian | 1078 | 1090 |
| LMArena Spanish | 1064 | 1111 |
Instruction Following DBRX leads
DBRX: 57.5 (#270), Mixtral 8x7B: 51.0 (#297)
| Benchmark | DBRX | Mixtral 8x7B |
|---|---|---|
| LMArena Instruction Following | 1112 | 1109 |
| IFEval | — | 57.5% |
Long Context Too close to call
DBRX: 33.7 (#258), Mixtral 8x7B: 33.4 (#260)
| Benchmark | DBRX | Mixtral 8x7B |
|---|---|---|
| LMArena Longer Query | 1112 | 1103 |
Writing & Preference Too close to call
DBRX: 33.6 (#275), Mixtral 8x7B: 34.2 (#270)
| Benchmark | DBRX | Mixtral 8x7B |
|---|---|---|
| LMArena Text | 1119 | 1132 |
| LMArena Creative Writing | 1104 | 1109 |
| LMArena Multi-Turn | 1111 | 1115 |
| WildBench | — | 67.3% |
Frequently asked questions
Is DBRX better than Mixtral 8x7B?
DBRX is the stronger model overall, scoring 29.4 to 27.1 on the Noometry Index.
Is DBRX or Mixtral 8x7B better for coding?
They score almost the same on coding (32.9 vs 32.8); test both on your own repository before choosing.
How many benchmarks do DBRX and Mixtral 8x7B share?
21 benchmarks have published results for both models. DBRX has 21 scored results on Noometry and Mixtral 8x7B has 38.