Model comparison
Codellama 70b Instruct vs DBRX
Codellama 70b Instruct is the stronger model overall, scoring 33.7 to 29.4 on the Noometry Index.
Last verified . 5 shared benchmarks.
Meta
DatabricksSummary
- They share 5 benchmarks with published results for both. Codellama 70b Instruct scores higher in 1 category and DBRX in 4 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where DBRX leads 57.5 to 51.9.
Side by side
| Codellama 70b Instruct | DBRX | |
|---|---|---|
| Provider | Meta | Databricks |
| Noometry Index | 33.7 | 29.4 |
| Released | — | 2024-03-27 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 7 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Codellama 70b Instruct leads
Codellama 70b Instruct: 37.6 (#193), DBRX: 32.9 (#266)
| Benchmark | Codellama 70b Instruct | DBRX |
|---|---|---|
| HumanEval+ | 65.9% | 70.1% |
| BigCodeBench Instruct | 40.7% | — |
| LMArena Coding | — | 1132 |
| BigCodeBench Complete | 49.6% | — |
| MBPP+ | — | 55.8% |
Reasoning DBRX leads
Codellama 70b Instruct: 20.1 (#242), DBRX: 21.4 (#222)
| Benchmark | Codellama 70b Instruct | DBRX |
|---|---|---|
| LMArena Hard Prompts | 1052 | 1113 |
Math Not comparable
Codellama 70b Instruct: —, DBRX: 24.3 (#269)
| Benchmark | Codellama 70b Instruct | DBRX |
|---|---|---|
| LMArena Math | — | 1145 |
| MATH Level 5 | — | 11.7% |
Knowledge Not comparable
Codellama 70b Instruct: —, DBRX: 14.9 (#294)
| Benchmark | Codellama 70b Instruct | DBRX |
|---|---|---|
| GPQA Diamond | — | 32.9% |
| LMArena Expert | — | 1076 |
Multilingual DBRX leads
Codellama 70b Instruct: 24.8 (#288), DBRX: 29.3 (#268)
| Benchmark | Codellama 70b Instruct | DBRX |
|---|---|---|
| LMArena Non-English | 992 | 1071 |
| LMArena Chinese | — | 1068 |
| LMArena French | — | 1096 |
| LMArena German | — | 1057 |
| LMArena Japanese | — | 990 |
| LMArena Korean | — | 993 |
| LMArena Russian | — | 1078 |
| LMArena Spanish | — | 1064 |
Instruction Following DBRX leads
Codellama 70b Instruct: 51.9 (#293), DBRX: 57.5 (#270)
| Benchmark | Codellama 70b Instruct | DBRX |
|---|---|---|
| LMArena Instruction Following | 1024 | 1112 |
Long Context Not comparable
Codellama 70b Instruct: —, DBRX: 33.7 (#258)
| Benchmark | Codellama 70b Instruct | DBRX |
|---|---|---|
| LMArena Longer Query | — | 1112 |
Writing & Preference Too close to call
Codellama 70b Instruct: 33.4 (#277), DBRX: 33.6 (#275)
| Benchmark | Codellama 70b Instruct | DBRX |
|---|---|---|
| LMArena Text | 1057 | 1119 |
| LMArena Creative Writing | — | 1104 |
| LMArena Multi-Turn | — | 1111 |
Frequently asked questions
Is Codellama 70b Instruct better than DBRX?
Codellama 70b Instruct is the stronger model overall, scoring 33.7 to 29.4 on the Noometry Index.
Is Codellama 70b Instruct or DBRX better for coding?
Codellama 70b Instruct scores higher on coding benchmarks: 37.6 versus 32.9 in the Noometry coding category.
How many benchmarks do Codellama 70b Instruct and DBRX share?
5 benchmarks have published results for both models. Codellama 70b Instruct has 7 scored results on Noometry and DBRX has 21.