Model comparison
Codestral vs StarCoder 2 7B
Codestral has enough public results to be ranked (#290); StarCoder 2 7B does not yet, so treat this comparison as directional.
Last verified . 2 shared benchmarks.
Mistral AI
NVIDIASummary
- They share 2 benchmarks with published results for both. Codestral scores higher in 0 categories and StarCoder 2 7B in 1 category; one gap is clear of the uncertainty.
- The widest gap is in coding, where StarCoder 2 7B leads 34.8 to 27.3.
- The biggest single-benchmark swing is BigCodeBench Complete: 52.5% for Codestral and 27.7% for StarCoder 2 7B.
- StarCoder 2 7B has downloadable open weights; the other is API-only.
Side by side
| Codestral | StarCoder 2 7B | |
|---|---|---|
| Provider | Mistral AI | NVIDIA |
| Noometry Index | 30.6 | 37.0 |
| Released | 2024-05-29 | 2024-02-20 |
| Weights | Proprietary | Open |
| Context window | 256K | — |
| Max output | 8K | — |
| Input $ / M tokens | $0.30 | — |
| Output $ / M tokens | $0.90 | — |
| Results tracked | 7 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding StarCoder 2 7B leads
Codestral: 27.3 (#321), StarCoder 2 7B: 34.8
| Benchmark | Codestral | StarCoder 2 7B |
|---|---|---|
| BigCodeBench Complete | 52.5% | 27.7% |
| HumanEval+ | 73.8% | 29.9% |
| Aider Polyglot | 11.1% | — |
| BigCodeBench Instruct | 41.8% | — |
| ALE-Bench | 137.78 | — |
| MBPP+ | 61.9% | — |
Reasoning Not comparable
Codestral: 19.8 (#251), StarCoder 2 7B: —
| Benchmark | Codestral | StarCoder 2 7B |
|---|---|---|
| Kagi LLM Benchmark | 32.5% | — |
| Epoch Capabilities Index | — | 93.56 |
| WinoGrande | — | 57.1% |
Math Not comparable
Codestral: —, StarCoder 2 7B: —
| Benchmark | Codestral | StarCoder 2 7B |
|---|---|---|
| GSM8K | — | 32.7% |
Knowledge Not comparable
Codestral: —, StarCoder 2 7B: —
| Benchmark | Codestral | StarCoder 2 7B |
|---|---|---|
| ARC (AI2) Challenge | — | 38.7% |
| MMLU | — | 38.8% |
Frequently asked questions
Is Codestral better than StarCoder 2 7B?
Codestral has enough public results to be ranked (#290); StarCoder 2 7B does not yet, so treat this comparison as directional.
Is Codestral or StarCoder 2 7B better for coding?
StarCoder 2 7B scores higher on coding benchmarks: 34.8 versus 27.3 in the Noometry coding category.
How many benchmarks do Codestral and StarCoder 2 7B share?
2 benchmarks have published results for both models. Codestral has 7 scored results on Noometry and StarCoder 2 7B has 7.