Side by side
See the difference.
Scores, capabilities and costs for the models you’re considering.
| 1 model | QwenQwen2.5-72B-InstructRemove × |
|---|---|
| What’s different | Qwen2.5 is a model pretrained on a large-scale dataset of up to 18 trillion tokens, offering significant improvements in knowledge, coding, mathematics, and instruction following compared to its predecessor… |
| Date | Sep 16, 2024Listed |
API prices USD · provider configuration shown below | |
| Listed rate | $0.36 in / $0.40 out/ 1M tokensdeepinfra · via Hugging Face · all routes ↗ |
| Provider endpoints | 5 endpoints → |
Specifications | |
| Context window | 32.8K |
| Max output | 16.4K |
| Parameters | 72B |
| Tool calling | Supported |
| Structured output | Supported |
| Open weights | Supported |