self-hosted/ai
§01·model · /models

Qwen3 32B

llmactive
§02·GPUs that run this model
4 total
GPUVRAMSeriesBest speedMin VRAMWorksBenchmarks
RTX 509032GB5061.4tokens/s3check ↗
RTX 3090 Ti24GB3038tokens/s24GB2check ↗
RTX 309024GB3035.1tokens/s24GB2check ↗
Apple M4 Max0GBapple35tokens/s20GB1check ↗