Model comparison
DBRX vs GPT-4 Turbo
GPT-4 Turbo is the stronger model overall, scoring 30.3 to 29.0 on the Noometry Index.
Last verified . 21 shared benchmarks.
DatabricksSummary
- They share 21 benchmarks with published results for both. DBRX scores higher in 2 categories and GPT-4 Turbo in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-4 Turbo leads 47.9 to 33.7.
- The biggest single-benchmark swing is MATH Level 5: 11.7% for DBRX and 46.7% for GPT-4 Turbo.
- DBRX has downloadable open weights; the other is API-only.
Side by side
| DBRX | GPT-4 Turbo | |
|---|---|---|
| Provider | Databricks | OpenAI |
| Noometry Index | 29.0 | 30.3 |
| Released | 2024-03-27 | 2023-11-06 |
| Weights | Open | Proprietary |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $10 |
| Output $ / M tokens | — | $30 |
| Results tracked | 21 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-4 Turbo leads
DBRX: 32.3 (#262), GPT-4 Turbo: 33.4 (#247)
| Benchmark | DBRX | GPT-4 Turbo |
|---|---|---|
| LMArena Coding | 1132 | 1268 |
| HumanEval+ | 70.1% | 86.6% |
| MBPP+ | 55.8% | 73.3% |
| WeirdML | — | 18% |
| BigCodeBench Instruct | — | 48.2% |
| BigCodeBench Complete | — | 58.2% |
Agentic & Tool Use Not comparable
DBRX: —, GPT-4 Turbo: —
| Benchmark | DBRX | GPT-4 Turbo |
|---|---|---|
| METR Time Horizons | — | 36.7% |
Reasoning DBRX leads
DBRX: 20.5 (#218), GPT-4 Turbo: 14.9 (#311)
| Benchmark | DBRX | GPT-4 Turbo |
|---|---|---|
| LMArena Hard Prompts | 1113 | 1251 |
| SimpleBench | — | 25.1% |
| Chess Puzzles | — | 6% |
| DTBench | — | 61.6% |
| LMCA | — | 9.8% |
| Epoch Capabilities Index | — | 127.25 |
| ForecastBench | — | 59.4 |
Math DBRX leads
DBRX: 22.3 (#265), GPT-4 Turbo: 8.6 (#316)
| Benchmark | DBRX | GPT-4 Turbo |
|---|---|---|
| LMArena Math | 1145 | 1272 |
| MATH Level 5 | 11.7% | 46.7% |
| FrontierMath (Tiers 1-3) | — | 0.7% |
| OTIS Mock AIME 2024-2025 | — | 6.7% |
Knowledge GPT-4 Turbo leads
DBRX: 14.9 (#289), GPT-4 Turbo: 24.3 (#263)
| Benchmark | DBRX | GPT-4 Turbo |
|---|---|---|
| GPQA Diamond | 32.9% | 46.6% |
| LMArena Expert | 1076 | 1223 |
| Confabulations | — | 28.4% |
| MMLU | — | 81.3% |
Multimodal Not comparable
DBRX: —, GPT-4 Turbo: 30.0 (#104)
| Benchmark | DBRX | GPT-4 Turbo |
|---|---|---|
| LMArena Vision | — | 1090 |
Multilingual GPT-4 Turbo leads
DBRX: 29.4 (#263), GPT-4 Turbo: 40.7 (#211)
| Benchmark | DBRX | GPT-4 Turbo |
|---|---|---|
| LMArena Non-English | 1071 | 1245 |
| LMArena Chinese | 1068 | 1242 |
| LMArena French | 1095 | 1275 |
| LMArena German | 1057 | 1259 |
| LMArena Japanese | 990 | 1194 |
| LMArena Korean | 993 | 1187 |
| LMArena Russian | 1078 | 1259 |
| LMArena Spanish | 1064 | 1260 |
Instruction Following GPT-4 Turbo leads
DBRX: 57.7 (#265), GPT-4 Turbo: 66.0 (#211)
| Benchmark | DBRX | GPT-4 Turbo |
|---|---|---|
| LMArena Instruction Following | 1112 | 1250 |
Long Context GPT-4 Turbo leads
DBRX: 33.7 (#253), GPT-4 Turbo: 38.0 (#200)
| Benchmark | DBRX | GPT-4 Turbo |
|---|---|---|
| LMArena Longer Query | 1112 | 1254 |
Writing & Preference GPT-4 Turbo leads
DBRX: 33.7 (#271), GPT-4 Turbo: 47.9 (#201)
| Benchmark | DBRX | GPT-4 Turbo |
|---|---|---|
| LMArena Text | 1119 | 1272 |
| LMArena Creative Writing | 1104 | 1269 |
| LMArena Multi-Turn | 1111 | 1267 |
Frequently asked questions
Is DBRX better than GPT-4 Turbo?
GPT-4 Turbo is the stronger model overall, scoring 30.3 to 29.0 on the Noometry Index.
Is DBRX or GPT-4 Turbo better for coding?
GPT-4 Turbo scores higher on coding benchmarks: 33.4 versus 32.3 in the Noometry coding category.
How many benchmarks do DBRX and GPT-4 Turbo share?
21 benchmarks have published results for both models. DBRX has 21 scored results on Noometry and GPT-4 Turbo has 36.