Model comparison
Codestral vs Command R+
Command R+ is the stronger model overall, scoring 32.4 to 30.6 on the Noometry Index. Codestral costs 9.7× less per token, which makes it the better buy when Command R+'s lead doesn't matter for your workload.
Last verified . 4 shared benchmarks.
Mistral AI
CohereSummary
- They share 4 benchmarks with published results for both. Codestral scores higher in 1 category and Command R+ in 1 category; 2 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Codestral leads 19.8 to 9.2.
- The biggest single-benchmark swing is BigCodeBench Complete: 52.5% for Codestral and 41.9% for Command R+.
- Codestral is cheaper at $0.30 / $0.90 per million input/output tokens, against $2.50 / $10 for Command R+.
- Codestral accepts more context: 256K tokens versus 128K.
- Command R+ has downloadable open weights; the other is API-only.
Side by side
| Codestral | Command R+ | |
|---|---|---|
| Provider | Mistral AI | Cohere |
| Noometry Index | 30.6 | 32.4 |
| Released | 2024-05-29 | 2024-08-30 |
| Weights | Proprietary | Open |
| Context window | 256K | 128K |
| Max output | 8K | 4K |
| Input $ / M tokens | $0.30 | $2.50 |
| Output $ / M tokens | $0.90 | $10 |
| Results tracked | 7 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Command R+ leads
Codestral: 27.3 (#321), Command R+: 29.1 (#309)
| Benchmark | Codestral | Command R+ |
|---|---|---|
| BigCodeBench Instruct | 41.8% | 33.8% |
| BigCodeBench Complete | 52.5% | 41.9% |
| HumanEval+ | 73.8% | 56.7% |
| MBPP+ | 61.9% | 63.5% |
| Aider Polyglot | 11.1% | — |
| LiveBench Coding | — | 19.1% |
| LMArena Coding | — | 1187 |
| ALE-Bench | 137.78 | — |
Reasoning Codestral leads
Codestral: 19.8 (#251), Command R+: 9.2 (#344)
| Benchmark | Codestral | Command R+ |
|---|---|---|
| SimpleBench | — | 17.4% |
| Kagi LLM Benchmark | 32.5% | — |
| LiveBench Reasoning | — | 24.8% |
| LMArena Hard Prompts | — | 1186 |
| DTBench | — | 54.9% |
| LiveBench Data Analysis | — | 38.1% |
| LMCA | — | 5% |
| Epoch Capabilities Index | — | 119.34 |
| LiveBench | — | 31.8% |
Math Not comparable
Codestral: —, Command R+: 28.9 (#242)
| Benchmark | Codestral | Command R+ |
|---|---|---|
| LiveBench Math | — | 21.3% |
| LMArena Math | — | 1188 |
Knowledge Not comparable
Codestral: —, Command R+: 36.4 (#169)
| Benchmark | Codestral | Command R+ |
|---|---|---|
| Vectara Hallucination Rate | — | 6.9% |
| LMArena Expert | — | 1174 |
| MMLU | — | 69.4% |
Multilingual Not comparable
Codestral: —, Command R+: 38.6 (#227)
| Benchmark | Codestral | Command R+ |
|---|---|---|
| LMArena Non-English | — | 1216 |
| LMArena Chinese | — | 1226 |
| LMArena French | — | 1209 |
| LMArena German | — | 1216 |
| LMArena Japanese | — | 1166 |
| LMArena Korean | — | 1138 |
| LMArena Russian | — | 1227 |
| LMArena Spanish | — | 1189 |
Instruction Following Not comparable
Codestral: —, Command R+: 60.0 (#254)
| Benchmark | Codestral | Command R+ |
|---|---|---|
| LiveBench Instruction Following | — | 57.6% |
| LMArena Instruction Following | — | 1197 |
Long Context Not comparable
Codestral: —, Command R+: 37.3 (#219)
| Benchmark | Codestral | Command R+ |
|---|---|---|
| LMArena Longer Query | — | 1230 |
Writing & Preference Not comparable
Codestral: —, Command R+: 43.5 (#228)
| Benchmark | Codestral | Command R+ |
|---|---|---|
| LMArena Text | — | 1229 |
| LMArena Creative Writing | — | 1235 |
| LMArena Multi-Turn | — | 1213 |
| LiveBench Language | — | 29.7% |
Frequently asked questions
Is Codestral better than Command R+?
Command R+ is the stronger model overall, scoring 32.4 to 30.6 on the Noometry Index. Codestral costs 9.7× less per token, which makes it the better buy when Command R+'s lead doesn't matter for your workload.
Which is cheaper, Codestral or Command R+?
Codestral is cheaper. It lists at $0.30 per million input tokens and $0.90 per million output tokens; Command R+ lists at $2.50 and $10.
Is Codestral or Command R+ better for coding?
Command R+ scores higher on coding benchmarks: 29.1 versus 27.3 in the Noometry coding category.
Which has the bigger context window?
Codestral does, with 256K tokens against 128K.
How many benchmarks do Codestral and Command R+ share?
4 benchmarks have published results for both models. Codestral has 7 scored results on Noometry and Command R+ has 34.