Model comparison
Codestral vs Gemma 2 27B
Codestral is the stronger model overall, scoring 30.6 to 29.4 on the Noometry Index.
Last verified . 2 shared benchmarks.
Mistral AI
GoogleSummary
- They share 2 benchmarks with published results for both. Codestral scores higher in 1 category and Gemma 2 27B in 1 category; 2 gaps are clear of the uncertainty.
- The widest gap is in coding, where Gemma 2 27B leads 34.1 to 27.3.
- Codestral is cheaper at $0.30 / $0.90 per million input/output tokens, against $0.65 / $0.65 for Gemma 2 27B.
- Codestral accepts more context: 256K tokens versus 8K.
- Gemma 2 27B has downloadable open weights; the other is API-only.
Side by side
| Codestral | Gemma 2 27B | |
|---|---|---|
| Provider | Mistral AI | |
| Noometry Index | 30.6 | 29.4 |
| Released | 2024-05-29 | 2024-06-24 |
| Weights | Proprietary | Open |
| Context window | 256K | 8K |
| Max output | 8K | 2K |
| Input $ / M tokens | $0.30 | $0.65 |
| Output $ / M tokens | $0.90 | $0.65 |
| Results tracked | 7 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 2 27B leads
Codestral: 27.3 (#321), Gemma 2 27B: 34.1 (#246)
| Benchmark | Codestral | Gemma 2 27B |
|---|---|---|
| BigCodeBench Instruct | 41.8% | 42.8% |
| BigCodeBench Complete | 52.5% | 52.5% |
| Aider Polyglot | 11.1% | — |
| LiveBench Coding | — | 36% |
| LMArena Coding | — | 1211 |
| ALE-Bench | 137.78 | — |
| HumanEval+ | 73.8% | — |
| MBPP+ | 61.9% | — |
Reasoning Codestral leads
Codestral: 19.8 (#251), Gemma 2 27B: 15.3 (#315)
| Benchmark | Codestral | Gemma 2 27B |
|---|---|---|
| Kagi LLM Benchmark | 32.5% | — |
| LiveBench Reasoning | — | 28.1% |
| LMArena Hard Prompts | — | 1198 |
| DTBench | — | 48% |
| LiveBench Data Analysis | — | 47.9% |
| LMCA | — | 7.1% |
| Epoch Capabilities Index | — | 122.08 |
| LiveBench | — | 38.2% |
Math Not comparable
Codestral: —, Gemma 2 27B: 10.7 (#311)
| Benchmark | Codestral | Gemma 2 27B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 1.4% |
| LiveBench Math | — | 26.5% |
| LMArena Math | — | 1212 |
| MATH Level 5 | — | 27.9% |
Knowledge Not comparable
Codestral: —, Gemma 2 27B: 19.0 (#280)
| Benchmark | Codestral | Gemma 2 27B |
|---|---|---|
| GPQA Diamond | — | 36.5% |
| Confabulations | — | 27.1% |
| LMArena Expert | — | 1172 |
| MMLU | — | 75.7% |
Multilingual Not comparable
Codestral: —, Gemma 2 27B: 38.6 (#226)
| Benchmark | Codestral | Gemma 2 27B |
|---|---|---|
| LMArena Non-English | — | 1217 |
| LMArena Chinese | — | 1221 |
| LMArena French | — | 1247 |
| LMArena German | — | 1209 |
| LMArena Japanese | — | 1175 |
| LMArena Korean | — | 1174 |
| LMArena Russian | — | 1234 |
| LMArena Spanish | — | 1228 |
Instruction Following Not comparable
Codestral: —, Gemma 2 27B: 60.5 (#249)
| Benchmark | Codestral | Gemma 2 27B |
|---|---|---|
| LiveBench Instruction Following | — | 58.1% |
| LMArena Instruction Following | — | 1206 |
Long Context Not comparable
Codestral: —, Gemma 2 27B: 37.3 (#218)
| Benchmark | Codestral | Gemma 2 27B |
|---|---|---|
| LMArena Longer Query | — | 1231 |
Writing & Preference Not comparable
Codestral: —, Gemma 2 27B: 44.2 (#225)
| Benchmark | Codestral | Gemma 2 27B |
|---|---|---|
| LMArena Text | — | 1231 |
| LMArena Creative Writing | — | 1241 |
| LMArena Multi-Turn | — | 1224 |
| LiveBench Language | — | 32.6% |
Frequently asked questions
Is Codestral better than Gemma 2 27B?
Codestral is the stronger model overall, scoring 30.6 to 29.4 on the Noometry Index.
Which is cheaper, Codestral or Gemma 2 27B?
Codestral is cheaper. It lists at $0.30 per million input tokens and $0.90 per million output tokens; Gemma 2 27B lists at $0.65 and $0.65.
Is Codestral or Gemma 2 27B better for coding?
Gemma 2 27B scores higher on coding benchmarks: 34.1 versus 27.3 in the Noometry coding category.
Which has the bigger context window?
Codestral does, with 256K tokens against 8K.
How many benchmarks do Codestral and Gemma 2 27B share?
2 benchmarks have published results for both models. Codestral has 7 scored results on Noometry and Gemma 2 27B has 34.