self-hosted/ai
§01·model · /models

Llama 3.1 34B

llmactive
§02·GPUs that run this model
5 total
GPUVRAMSeriesBest speedMin VRAMWorksBenchmarks
RTX 409024GB4042tok/s24GB1check ↗
RTX 309024GB3028tok/s24GB1check ↗
Apple M3 Max0GBapple11tok/s64GB1check ↗
RTX 4060 Ti 16GB16GB408tok/s16GB1check ↗
RTX 4070 Super12GB4012GB1check ↗