self-hosted/ai
§01·model · /models

Llama 3.1 70B

llmactive
§02·GPUs that run this model
4 total
GPUVRAMSeriesBest speedMin VRAMWorksBenchmarks
RTX 409024GB4018tok/s24GB1check ↗
RTX 309024GB3010tok/s24GB1check ↗
Apple M3 Max0GBapple5tok/s64GB1check ↗
RTX 4060 Ti 16GB16GB4016GB1check ↗