Open-weight LLMs you can run yourself › Qwen2.5 7B
Updated · public data
Alibaba · Open weights
Qwen2.5 7B is an open-weight LLM from Alibaba with 7.6B parameters. It needs about 5 GB of memory at 4-bit — smallest setups that run it: PC with a GPU 8 GB, PC or laptop, no GPU 16 GB, Mac 16 GB.
| Parameters | 7.6B |
| Memory to run | ~5 GB (at 4-bit, rough estimate) |
| Downloads | 8M (last 30 days on Hugging Face) |
| Likes | 2.4K |
| License | Apache 2.0 |
| First uploaded | Sep 15, 2024 |
| Hosted from | $0.1 (per 1M input tokens · cheapest on OpenRouter) |
| Phone | iPhone · Android: too big |
| PC with a GPU | 8 GB: fits · 12 GB: fits · 16 GB: fits · 24 GB: fits · 32 GB: fits |
| PC or laptop, no GPU | 16 GB: fits · 32 GB: fits · 64 GB: fits |
| Mac | 16 GB: fits · 32 GB: fits · 64 GB: fits · 128 GB: fits |
huggingface.co/Qwen/Qwen2.5-7B-Instruct
Data: Hugging Face (official accounts), Oct 16, 2026 · memory is a rough estimate (4-bit for LLMs, 8-bit for image and video, plus 10%) · hosted price: OpenRouter