Model comparison
DBRX vs GPT-4
DBRX and GPT-4 score almost the same on the Noometry Index (29.4 vs 29.1), so choose on price, context window or the category you care about most.
Last verified . 20 shared benchmarks.
DatabricksSummary
- They share 20 benchmarks with published results for both. DBRX scores higher in 3 categories and GPT-4 in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where DBRX leads 24.3 to 10.8.
- The biggest single-benchmark swing is MATH Level 5: 11.7% for DBRX and 23% for GPT-4.
- DBRX has downloadable open weights; the other is API-only.
Side by side
| DBRX | GPT-4 | |
|---|---|---|
| Provider | Databricks | OpenAI |
| Noometry Index | 29.4 | 29.1 |
| Released | 2024-03-27 | 2023-03-14 |
| Weights | Open | Proprietary |
| Context window | — | 8K |
| Max output | — | 8K |
| Input $ / M tokens | — | $30 |
| Output $ / M tokens | — | $60 |
| Results tracked | 21 | 38 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DBRX leads
DBRX: 32.9 (#266), GPT-4: 31.6 (#283)
| Benchmark | DBRX | GPT-4 |
|---|---|---|
| LMArena Coding | 1132 | 1254 |
| HumanEval+ | 70.1% | 79.3% |
| WeirdML | — | 12.4% |
| BigCodeBench Instruct | — | 46% |
| BigCodeBench Complete | — | 57.2% |
| MBPP+ | 55.8% | — |
Agentic & Tool Use Not comparable
DBRX: —, GPT-4: —
| Benchmark | DBRX | GPT-4 |
|---|---|---|
| METR Time Horizons | — | 36.1% |
Reasoning DBRX leads
DBRX: 21.4 (#222), GPT-4: 17.8 (#289)
| Benchmark | DBRX | GPT-4 |
|---|---|---|
| LMArena Hard Prompts | 1113 | 1241 |
| Chess Puzzles | — | 4% |
| Mystery Game Puzzles | — | 12% |
| DTBench | — | 62.7% |
| LMCA | — | 17.1% |
| BIG-Bench Hard | — | 75.1% |
| Epoch Capabilities Index | — | 125.89 |
| ForecastBench | — | 57.8 |
| HellaSwag | — | 95.3% |
| WinoGrande | — | 87.5% |
Math DBRX leads
DBRX: 24.3 (#269), GPT-4: 10.8 (#309)
| Benchmark | DBRX | GPT-4 |
|---|---|---|
| LMArena Math | 1145 | 1269 |
| MATH Level 5 | 11.7% | 23% |
| OTIS Mock AIME 2024-2025 | — | 1.1% |
| GSM8K | — | 92% |
Knowledge GPT-4 leads
DBRX: 14.9 (#294), GPT-4: 18.4 (#282)
| Benchmark | DBRX | GPT-4 |
|---|---|---|
| GPQA Diamond | 32.9% | 35.7% |
| LMArena Expert | 1076 | 1211 |
| MMLU | — | 86.4% |
| TriviaQA | — | 84.8% |
Multilingual GPT-4 leads
DBRX: 29.3 (#268), GPT-4: 40.6 (#215)
| Benchmark | DBRX | GPT-4 |
|---|---|---|
| LMArena Non-English | 1071 | 1246 |
| LMArena Chinese | 1068 | 1242 |
| LMArena French | 1096 | 1283 |
| LMArena German | 1057 | 1251 |
| LMArena Japanese | 990 | 1209 |
| LMArena Korean | 993 | 1184 |
| LMArena Russian | 1078 | 1251 |
| LMArena Spanish | 1064 | 1261 |
Instruction Following GPT-4 leads
DBRX: 57.5 (#270), GPT-4: 65.3 (#222)
| Benchmark | DBRX | GPT-4 |
|---|---|---|
| LMArena Instruction Following | 1112 | 1241 |
Long Context GPT-4 leads
DBRX: 33.7 (#258), GPT-4: 37.7 (#212)
| Benchmark | DBRX | GPT-4 |
|---|---|---|
| LMArena Longer Query | 1112 | 1244 |
Writing & Preference GPT-4 leads
DBRX: 33.6 (#275), GPT-4: 34.9 (#268)
| Benchmark | DBRX | GPT-4 |
|---|---|---|
| LMArena Text | 1119 | 1263 |
| LMArena Creative Writing | 1104 | 1244 |
| LMArena Multi-Turn | 1111 | 1257 |
| EQ-Bench Creative Writing | — | 752 |
Frequently asked questions
Is DBRX better than GPT-4?
DBRX and GPT-4 score almost the same on the Noometry Index (29.4 vs 29.1), so choose on price, context window or the category you care about most.
Is DBRX or GPT-4 better for coding?
DBRX scores higher on coding benchmarks: 32.9 versus 31.6 in the Noometry coding category.
How many benchmarks do DBRX and GPT-4 share?
20 benchmarks have published results for both models. DBRX has 21 scored results on Noometry and GPT-4 has 38.