Model comparison
Codellama 70b Instruct vs Gemma 7B
Codellama 70b Instruct is the stronger model overall, scoring 33.7 to 30.0 on the Noometry Index.
Last verified . 5 shared benchmarks.
Meta
GoogleSummary
- They share 5 benchmarks with published results for both. Codellama 70b Instruct scores higher in 4 categories and Gemma 7B in 1 category; 2 gaps are clear of the uncertainty.
- The widest gap is in coding, where Codellama 70b Instruct leads 37.6 to 30.5.
Side by side
| Codellama 70b Instruct | Gemma 7B | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 33.7 | 30.0 |
| Released | — | 2024-02-21 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 7 | 27 |
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), Gemma 7B: 30.5 (#294)
| Benchmark | Codellama 70b Instruct | Gemma 7B |
|---|---|---|
| HumanEval+ | 65.9% | 28.7% |
| BigCodeBench Instruct | 40.7% | — |
| LMArena Coding | — | 1048 |
| BigCodeBench Complete | 49.6% | — |
| MBPP+ | — | 43.4% |
Reasoning Too close to call
Codellama 70b Instruct: 20.1 (#242), Gemma 7B: 19.9 (#249)
| Benchmark | Codellama 70b Instruct | Gemma 7B |
|---|---|---|
| LMArena Hard Prompts | 1052 | 1042 |
| Adversarial NLI | — | 48.7% |
| BIG-Bench Hard | — | 55.1% |
| Epoch Capabilities Index | — | 111.99 |
| HellaSwag | — | 82.2% |
| PIQA | — | 81.2% |
| WinoGrande | — | 79% |
Math Not comparable
Codellama 70b Instruct: —, Gemma 7B: 31.2 (#228)
| Benchmark | Codellama 70b Instruct | Gemma 7B |
|---|---|---|
| LMArena Math | — | 1066 |
| GSM8K | — | 46.4% |
Knowledge Not comparable
Codellama 70b Instruct: —, Gemma 7B: 27.3 (#252)
| Benchmark | Codellama 70b Instruct | Gemma 7B |
|---|---|---|
| LMArena Expert | — | 1001 |
| ARC (AI2) Challenge | — | 78.3% |
| BoolQ | — | 83.2% |
| MMLU | — | 66.1% |
| OpenBookQA | — | 78.6% |
| TriviaQA | — | 72.3% |
Multilingual Too close to call
Codellama 70b Instruct: 24.8 (#288), Gemma 7B: 25.1 (#287)
| Benchmark | Codellama 70b Instruct | Gemma 7B |
|---|---|---|
| LMArena Non-English | 992 | 999 |
| LMArena Chinese | — | 1035 |
| LMArena French | — | 1025 |
| LMArena Russian | — | 993 |
Instruction Following Too close to call
Codellama 70b Instruct: 51.9 (#293), Gemma 7B: 51.5 (#295)
| Benchmark | Codellama 70b Instruct | Gemma 7B |
|---|---|---|
| LMArena Instruction Following | 1024 | 1017 |
Long Context Not comparable
Codellama 70b Instruct: —, Gemma 7B: 31.1 (#282)
| Benchmark | Codellama 70b Instruct | Gemma 7B |
|---|---|---|
| LMArena Longer Query | — | 1022 |
Writing & Preference Codellama 70b Instruct leads
Codellama 70b Instruct: 33.4 (#277), Gemma 7B: 27.1 (#302)
| Benchmark | Codellama 70b Instruct | Gemma 7B |
|---|---|---|
| LMArena Text | 1057 | 1056 |
| LMArena Creative Writing | — | 1024 |
| LMArena Multi-Turn | — | 963 |
Frequently asked questions
Is Codellama 70b Instruct better than Gemma 7B?
Codellama 70b Instruct is the stronger model overall, scoring 33.7 to 30.0 on the Noometry Index.
Is Codellama 70b Instruct or Gemma 7B better for coding?
Codellama 70b Instruct scores higher on coding benchmarks: 37.6 versus 30.5 in the Noometry coding category.
How many benchmarks do Codellama 70b Instruct and Gemma 7B share?
5 benchmarks have published results for both models. Codellama 70b Instruct has 7 scored results on Noometry and Gemma 7B has 27.