All Comparisons
Codestral-22B vs GPT-4o
Codestral-22B
Mistral AI 🇫🇷
GPT-4o
OpenAI 🇺🇸
58.1
#150—
high$2.50/M in · $10.00/M out
128K context
Most Affordable
GPT-4o
$2.50/M input
Largest Context
GPT-4o
128K tokens
Best Benchmark Score
GPT-4o
58.1/100
Model Specifications
| Spec | Codestral-22B | GPT-4o |
|---|---|---|
| Provider | Mistral AI | OpenAI |
| Release Date | May 2024 | May 2024 |
| Knowledge Cutoff | — | Oct 2023 |
| Parameters | 22200000000B | unknown |
| Context Window | — | 128K |
| 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 | GPT-4o |
|---|---|---|
| Input | — | $2.50 |
| Output | — | $10.00 |
API Provider Pricing
Capabilities
| Feature | Codestral-22B | GPT-4o |
|---|---|---|
| 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
general
| Benchmark | Codestral-22B | GPT-4o |
|---|---|---|
| MMLU-Pro | — | 74.5 |
| IFEval | — | 83.6 |
| Chatbot Arena Elo | — | 1285 |
coding
| Benchmark | Codestral-22B | GPT-4o |
|---|---|---|
| HumanEval | — | 90.2 |
| SWE-bench Verified | — | 33.2 |
reasoning
| Benchmark | Codestral-22B | GPT-4o |
|---|---|---|
| GPQA Diamond | — | 53.6 |
| ARC-Challenge | — | 96.4 |
| BigBench-Hard | — | 87.3 |
language
| Benchmark | Codestral-22B | GPT-4o |
|---|---|---|
| HellaSwag | — | 96.4 |
| WinoGrande | — | 85.7 |
multimodal
| Benchmark | Codestral-22B | GPT-4o |
|---|---|---|
| MMMU | — | 69.1 |
safety
| Benchmark | Codestral-22B | GPT-4o |
|---|---|---|
| TruthfulQA | — | 73.5 |
Category Winners
coding
38.2
math
54.5
reasoning
54.2
general
70.5
language
79.9
multimodal
52.9
safety
94.3