self-hosted/ai
§01·model · /models

Llama 3.1 7B

llmactive
§02·GPUs that run this model
6 total
GPUVRAMSeriesBest speedMin VRAMWorksBenchmarks
RTX 409024GB40135tok/s24GB1check ↗
RTX 309024GB3095tok/s24GB1check ↗
RTX 4070 Super12GB4075tok/s12GB1check ↗
RTX 4060 Ti 16GB16GB4055tok/s16GB1check ↗
RTX 306012GB3045tok/s12GB1check ↗
Apple M3 Max0GBapple40tok/s64GB1check ↗