Mistral AI model product
Codestral 22B is a 22B parameter code generation model trained on more than 80 programming languages.
Updated Sep 8, 2026. Default version: Codestral-22B
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Benchmark scores for Codestral-22B.
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkCruxEval-O | Score51.30% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkHumanEval-Average | Score61.50% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkHumanEvalFIM-Average | Score91.60% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkRepoBench | Score34.00% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkSpider | Score63.50% | Rank01 | Participants2 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkMBPP | Score78.20% | Rank15 | Participants33 | Percentile56.25% | EvidenceC | Evaluated |
| BenchmarkHumanEval | Score81.10% | Rank43 | Participants66 | Percentile35.38% | EvidenceC | Evaluated |
Preference and agent-evaluation results for the default version.
The default version does not have a matching Arena result yet.
Official vendor API pricing appears first, followed by individual provider offers.
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderPioneer | Provider model IDmistralai/Codestral-22B-v0.1 | Regionglobal | Input / 1M$0.30 | Output / 1M$0.90 | Context128K | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Provider-specific output speed and catalog latency for Codestral-22B. Runtime does not affect the capability score.
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Technical details for the model's default version.
Available versions of this model. The score column identifies the version used in the overall ranking.
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionCodestral-22B | Released | LLMBoard0.64 | Parameters22.2B | ContextN/A | Max outputN/A | Open weightsYes | LicenseMNPL-0.1 |
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
Key information about Codestral 22B and its available data.
Codestral 22B is a code generation model from Mistral AI trained on more than 80 programming languages, including Python, Java, C, C++, JavaScript, and Bash. It supports instruction-following and fill-in-the-middle (FIM) for code completion and generation tasks.
Data as of 2026-09-08.
Common questions about Codestral 22B.
Codestral 22B's default version was released on May 29, 2024.
No official standard PAYG price is currently available for Codestral 22B. The lowest tracked third-party offer starts at $0.30 input and $0.90 output via Pioneer.
Codestral 22B was created by Mistral AI.
A context window is not available for the default version.
Yes. The default version is marked as open weight under MNPL-0.1.
1 provider offerings are linked to the default version.