self-hosted/ai
§01·model · /models

Llama 3.1 70B

llmactiveLlama 3.1 Community License

70B instruction-tuned LLM by Meta, the Llama 3.1 generation, with 128K context. Coverage here is a single recipe on a 32 GB RTX 5090 — at this parameter count the list of consumer cards runs out quickly. Its successor, Llama 3.3 70B, is also catalogued and reaches four machines, three of them Apple. Llama 3.1 Community License.

§02·same family
1 other model

Other models grouped with Llama 3.1 70B. 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