Compare LLM Models
Select 2-5 models to compare side by side.
Select models to compare
0 / 5 selected
Top downloads
Select 2-5 models
| Attribute |
Qwen
|
|---|---|
| Company | Qwen |
| Release Date | 2025-09-23 |
| Parameters (B) | 235 |
| Architecture | MoE |
| Context Window | 256000 |
| Input Modalities | text, image |
| Open Source | Yes |
| License | apache-2.0 |
| Score | 0 |
| VRAM (GB) | 470 |
| Compute | Requires 8× NVIDIA A100/H100/H200 GPUs (80GB+ each) with tensor parallelism; ~470GB VRAM for BF16 inference |
| Benchmark: AIME-2025 | 74.7 |
| Benchmark: GPQA | 74.3 |
| Benchmark: IFEval | 87.8 |
| Benchmark: LiveCodeBench | 54.3 |
| Benchmark: MMLU-Pro | 81.8 |
| Benchmark: MMMU-Pro | 67.6 |
| Pricing (Input) | 0.21 |
| Pricing (Output) | 1.90 |