Model comparison
DBRX vs GPT-3.5-turbo
DBRX is the stronger model overall, scoring 29.4 to 23.2 on the Noometry Index.
Last verified . 21 shared benchmarks.
DatabricksSummary
- They share 21 benchmarks with published results for both. DBRX scores higher in 5 categories and GPT-3.5-turbo in 3 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where DBRX leads 24.3 to 6.3.
- DBRX has downloadable open weights; the other is API-only.
Side by side
| DBRX | GPT-3.5-turbo | |
|---|---|---|
| Provider | Databricks | OpenAI |
| Noometry Index | 29.4 | 23.2 |
| Released | 2024-03-27 | 2023-03-01 |
| Weights | Open | Proprietary |
| Context window | — | 16K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.50 |
| Output $ / M tokens | — | $1.50 |
| Results tracked | 21 | 44 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DBRX leads
DBRX: 32.9 (#266), GPT-3.5-turbo: 23.9 (#331)
| Benchmark | DBRX | GPT-3.5-turbo |
|---|---|---|
| LMArena Coding | 1132 | 1136 |
| HumanEval+ | 70.1% | 70.7% |
| MBPP+ | 55.8% | 69.7% |
| WeirdML | — | 3.5% |
| BigCodeBench Instruct | — | 39.1% |
| BigCodeBench Complete | — | 50.6% |
Agentic & Tool Use Not comparable
DBRX: —, GPT-3.5-turbo: —
| Benchmark | DBRX | GPT-3.5-turbo |
|---|---|---|
| METR Time Horizons | — | 21.5% |
Reasoning DBRX leads
DBRX: 21.4 (#222), GPT-3.5-turbo: 13.8 (#332)
| Benchmark | DBRX | GPT-3.5-turbo |
|---|---|---|
| LMArena Hard Prompts | 1113 | 1108 |
| Chess Puzzles | — | 0% |
| Mystery Game Puzzles | — | 3% |
| DTBench | — | 48.5% |
| LMCA | — | 9.7% |
| Adversarial NLI | — | 58.1% |
| BIG-Bench Hard | — | 61.6% |
| CommonsenseQA 2.0 | — | 57% |
| Epoch Capabilities Index | — | 118.55 |
| ForecastBench | — | 50.4 |
| WinoGrande | — | 81.6% |
Math DBRX leads
DBRX: 24.3 (#269), GPT-3.5-turbo: 6.3 (#327)
| Benchmark | DBRX | GPT-3.5-turbo |
|---|---|---|
| LMArena Math | 1145 | 1142 |
| MATH Level 5 | 11.7% | 15.9% |
| FrontierMath (Tiers 1-3) | — | 0% |
| OTIS Mock AIME 2024-2025 | — | 2.2% |
| GSM8K | — | 57.8% |
Knowledge DBRX leads
DBRX: 14.9 (#294), GPT-3.5-turbo: 10.0 (#303)
| Benchmark | DBRX | GPT-3.5-turbo |
|---|---|---|
| GPQA Diamond | 32.9% | 28% |
| LMArena Expert | 1076 | 1070 |
| ARC (AI2) Challenge | — | 87.4% |
| BoolQ | — | 87% |
| MMLU | — | 71.4% |
| OpenBookQA | — | 86% |
| TriviaQA | — | 85.8% |
Multilingual GPT-3.5-turbo leads
DBRX: 29.3 (#268), GPT-3.5-turbo: 31.5 (#258)
| Benchmark | DBRX | GPT-3.5-turbo |
|---|---|---|
| LMArena Non-English | 1071 | 1108 |
| LMArena Chinese | 1068 | 1075 |
| LMArena French | 1096 | 1118 |
| LMArena German | 1057 | 1090 |
| LMArena Japanese | 990 | 1043 |
| LMArena Korean | 993 | 1019 |
| LMArena Russian | 1078 | 1123 |
| LMArena Spanish | 1064 | 1121 |
Instruction Following Too close to call
DBRX: 57.5 (#270), GPT-3.5-turbo: 57.9 (#262)
| Benchmark | DBRX | GPT-3.5-turbo |
|---|---|---|
| LMArena Instruction Following | 1112 | 1119 |
Long Context Too close to call
DBRX: 33.7 (#258), GPT-3.5-turbo: 34.0 (#254)
| Benchmark | DBRX | GPT-3.5-turbo |
|---|---|---|
| LMArena Longer Query | 1112 | 1121 |
Writing & Preference DBRX leads
DBRX: 33.6 (#275), GPT-3.5-turbo: 25.3 (#305)
| Benchmark | DBRX | GPT-3.5-turbo |
|---|---|---|
| LMArena Text | 1119 | 1125 |
| LMArena Creative Writing | 1104 | 1092 |
| LMArena Multi-Turn | 1111 | 1117 |
| EQ-Bench Creative Writing | — | 451 |
Frequently asked questions
Is DBRX better than GPT-3.5-turbo?
DBRX is the stronger model overall, scoring 29.4 to 23.2 on the Noometry Index.
Is DBRX or GPT-3.5-turbo better for coding?
DBRX scores higher on coding benchmarks: 32.9 versus 23.9 in the Noometry coding category.
How many benchmarks do DBRX and GPT-3.5-turbo share?
21 benchmarks have published results for both models. DBRX has 21 scored results on Noometry and GPT-3.5-turbo has 44.