Mistral AI model product
Codestral 22B
A 22B parameter code generation model trained on 80+ programming languages including Python, Java, C, C++, JavaScript, and Bash.
Updated Aug 12, 2026. Default version: Codestral-22B
LLMBoard score1.6Codestral-22B
Coverage20%2 benchmark families
Context windowN/ATokens
Official input priceN/AOfficial price unavailable
Codestral 22B Specifications
Technical details for the model's default version.
- Version
- Codestral-22B
- Released
- May 29, 2024
- Knowledge cutoff
- Unknown
- Parameters
- 22.2B
- Context window
- N/A
- Max output
- N/A
- Inputs
- text
- Outputs
- text
- Open weights
- No
- License
- MNPL-0.1
Codestral 22B Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Codestral 22B Benchmark Results
Benchmark scores for Codestral-22B.
| CruxEval-O | 51.3% | 01 | 1 | 100.0% | C | |
| HumanEval-Average | 61.5% | 01 | 1 | 100.0% | C | |
| HumanEvalFIM-Average | 91.6% | 01 | 1 | 100.0% | C | |
| RepoBench | 34.0% | 01 | 1 | 100.0% | C | |
| Spider | 63.5% | 01 | 2 | 100.0% | C | |
| MBPP | 78.2% | 15 | 33 | 56.3% | C | |
| HumanEval | 81.1% | 43 | 66 | 35.4% | C |
Codestral 22B Arena Results
Preference and agent-evaluation results for the default version.
No Arena results
The default version does not have a matching Arena result yet.
Codestral 22B Runtime Performance
Provider-specific output speed and catalog latency for Codestral-22B. Runtime does not affect the capability score.
No runtime data
Browse runtime rankingsNo 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.
Codestral 22B Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- From $0.30 input, $0.90 output per 1M via Pioneer
- Tracked offerings
- 1
| Pioneer | mistralai/Codestral-22B-v0.1 | global | $0.30 | $0.90 | 128K |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Codestral 22B Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
| Codestral-22B | 1.6 | 22.2B | N/A | N/A | No | MNPL-0.1 |
What is Codestral 22B?
Key information about Codestral 22B and its available data.
A 22B parameter code generation model trained on 80+ programming languages including Python, Java, C, C++, JavaScript, and Bash. Supports both instruction-following and fill-in-the-middle (FIM) capabilities for code completion and generation tasks.
Data as of 2026-08-11.
Codestral 22B vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to Codestral 22B
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
MA
Mistral NeMo
Mistral AI
MA
Pixtral 12B
Mistral AI
MA
Ministral 8B
Mistral AI
MA
Mistral Large 2
Mistral AI
MA
Mistral Small 3 24B
Mistral AI
MA
Mistral Small 3.1 24B
Mistral AI
FAQ
Common questions about Codestral 22B.
When was Codestral 22B released?
Codestral 22B's default version was released on May 29, 2024.
How much does Codestral 22B cost?
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.
Who created Codestral 22B?
Codestral 22B was created by Mistral AI.
What is the context window for Codestral 22B?
A context window is not available for the default version.
Is Codestral 22B open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer Codestral 22B?
1 provider offerings are linked to the default version.
What models should I compare Codestral 22B with?
Nearby ranked alternatives include Llama 3.1 Nemotron 70B, Qwen3.5 2B, Jamba 1.5 Large.