self-hosted/ai
§01·model · /models

Qwen2.5 7B

llmactiveApache-2.0

7B instruction-tuned LLM by Alibaba, the Qwen2.5 generation — the step between Qwen2 and Qwen3, both of which are also catalogued here. One recipe covers it, on an 8 GB RTX 3060 Ti, where the shared 4-bit Ollama run records 58.13 tokens/s: slower than the Qwen2 it replaced on that same card, which is the usual trade when a generation adds capability rather than speed. Apache-2.0.

§02·same family
1 other model

Other models grouped with Qwen2.5 7B. Sizes and modalities can differ, and nothing here says whether one fits your GPU — open a model for its own compatibility table.

§03·GPUs that run this model
1 total

benchmarked·~ runs via recipe (not benchmarked)· untested·doesn't fit