Best Benchmark Score
o3
47.9/100
Model Specifications
| Spec | Codestral-22B | o3 |
|---|---|---|
| Provider | Mistral AI | OpenAI |
| Release Date | May 2024 | Apr 2025 |
| Knowledge Cutoff | — | 2024-05-31 |
| Parameters | 22200000000B | Undisclosed |
| Context Window | — | — |
| Max Output | — | — |
| Open Source | Yes | No |
| License | mnpl_0_1 | proprietary |
| Tokenizer | — | — |
| Modality | — | — |
| Reasoning Model | No | No |
| Moderated | No | No |
Pricing Comparison
| Price (per 1M tokens) | Codestral-22B | o3 |
|---|
API Provider Pricing
Capabilities
| Feature | Codestral-22B | o3 |
|---|---|---|
| Text Input | ||
| Image Input (Vision) | ||
| Audio Input | ||
| Video Input | ||
| File Input | ||
| Image Output | ||
| Audio Output | ||
| Tool Use | ||
| Structured Output (JSON) | ||
| Streaming | ||
| Reasoning Tokens | ||
| Web Search | ||
| Temperature Control | ||
| Top-P Sampling | ||
| Stop Sequences | ||
| Seed (Reproducibility) | ||
| Log Probabilities |
Category Comparison
Benchmark-by-Benchmark
coding
| Benchmark | Codestral-22B | o3 |
|---|---|---|
| SWE-bench Verified | — | 69.1 |
math
| Benchmark | Codestral-22B | o3 |
|---|---|---|
| AIME 2025 | — | 86.4 |
| FrontierMath | — | 15.8 |
reasoning
| Benchmark | Codestral-22B | o3 |
|---|---|---|
| GPQA Diamond | — | 83.3 |
| Humanity's Last Exam | — | 14.7 |
| ARC-AGI v2 | — | 6.5 |
multimodal
| Benchmark | Codestral-22B | o3 |
|---|---|---|
| MMMU | — | 82.9 |
| MMMU-Pro | — | 76.4 |
| CharXiv Reasoning | — | 78.6 |
agent
| Benchmark | Codestral-22B | o3 |
|---|---|---|
| BrowseComp | — | 49.7 |
Category Winners
coding
58.4
math
43.5
reasoning
34.1
multimodal
73.4
agent
23.4