Model comparison
Codestral vs Ministral 8B
Codestral is the stronger model overall, scoring 30.6 to 28.2 on the Noometry Index. Ministral 8B costs 3.0× less per token, which makes it the better buy when Codestral's lead doesn't matter for your workload.
Last verified . 0 shared benchmarks.
Mistral AISummary
- The widest gap is in coding, where Ministral 8B leads 35.0 to 27.3.
- Ministral 8B is cheaper at $0.15 / $0.15 per million input/output tokens, against $0.30 / $0.90 for Codestral.
- Ministral 8B accepts more context: 262K tokens versus 256K.
- Ministral 8B has downloadable open weights; the other is API-only.
Side by side
| Codestral | Ministral 8B | |
|---|---|---|
| Provider | Mistral AI | Mistral AI |
| Noometry Index | 30.6 | 28.2 |
| Released | 2024-05-29 | 2024-10-01 |
| Weights | Proprietary | Open |
| Context window | 256K | 262K |
| Max output | 8K | 262K |
| Input $ / M tokens | $0.30 | $0.15 |
| Output $ / M tokens | $0.90 | $0.15 |
| Results tracked | 7 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Ministral 8B leads
Codestral: 27.3 (#321), Ministral 8B: 35.0 (#230)
| Benchmark | Codestral | Ministral 8B |
|---|---|---|
| Aider Polyglot | 11.1% | — |
| BigCodeBench Instruct | 41.8% | — |
| LMArena Coding | — | 1202 |
| BigCodeBench Complete | 52.5% | — |
| ALE-Bench | 137.78 | — |
| HumanEval+ | 73.8% | — |
| MBPP+ | 61.9% | — |
Agentic & Tool Use Not comparable
Codestral: —, Ministral 8B: 16.4 (#148)
| Benchmark | Codestral | Ministral 8B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 11.1% |
Reasoning Codestral leads
Codestral: 19.8 (#251), Ministral 8B: 18.4 (#281)
| Benchmark | Codestral | Ministral 8B |
|---|---|---|
| Kagi LLM Benchmark | 32.5% | — |
| LMArena Hard Prompts | — | 1191 |
| DTBench | — | 45.7% |
Math Not comparable
Codestral: —, Ministral 8B: 25.7 (#267)
| Benchmark | Codestral | Ministral 8B |
|---|---|---|
| LMArena Math | — | 1188 |
| MATH Level 5 | — | 14.9% |
Knowledge Not comparable
Codestral: —, Ministral 8B: 12.6 (#297)
| Benchmark | Codestral | Ministral 8B |
|---|---|---|
| GPQA Diamond | — | 27.1% |
| Vectara Hallucination Rate | — | 7.4% |
| LMArena Expert | — | 1170 |
Multilingual Not comparable
Codestral: —, Ministral 8B: 35.1 (#247)
| Benchmark | Codestral | Ministral 8B |
|---|---|---|
| LMArena Non-English | — | 1165 |
| LMArena Chinese | — | 1193 |
| LMArena Russian | — | 1195 |
Instruction Following Not comparable
Codestral: —, Ministral 8B: 60.5 (#250)
| Benchmark | Codestral | Ministral 8B |
|---|---|---|
| LMArena Instruction Following | — | 1161 |
Long Context Not comparable
Codestral: —, Ministral 8B: 36.7 (#227)
| Benchmark | Codestral | Ministral 8B |
|---|---|---|
| LMArena Longer Query | — | 1212 |
Writing & Preference Not comparable
Codestral: —, Ministral 8B: 39.6 (#246)
| Benchmark | Codestral | Ministral 8B |
|---|---|---|
| LMArena Text | — | 1191 |
| LMArena Creative Writing | — | 1175 |
| LMArena Multi-Turn | — | 1166 |
Frequently asked questions
Is Codestral better than Ministral 8B?
Codestral is the stronger model overall, scoring 30.6 to 28.2 on the Noometry Index. Ministral 8B costs 3.0× less per token, which makes it the better buy when Codestral's lead doesn't matter for your workload.
Which is cheaper, Codestral or Ministral 8B?
Ministral 8B is cheaper. It lists at $0.15 per million input tokens and $0.15 per million output tokens; Codestral lists at $0.30 and $0.90.
Is Codestral or Ministral 8B better for coding?
Ministral 8B scores higher on coding benchmarks: 35.0 versus 27.3 in the Noometry coding category.
Which has the bigger context window?
Ministral 8B does, with 262K tokens against 256K.
How many benchmarks do Codestral and Ministral 8B share?
0 benchmarks have published results for both models. Codestral has 7 scored results on Noometry and Ministral 8B has 17.