vram.run Models Hardware Providers Cloud State of Inference
API provider data is live · Hardware & cloud pricing curated 2026-02-23

Qwen3-0.6B

by Qwen

752M params · text-generation · 1.1k likes · 12.2M downloads

Qwen3-0.6B is a 752M parameter model. At Q4 quantization it requires 0GB of VRAM. It runs comfortably on GeForce RTX 4090 (1746 tok/s), GeForce RTX 5090 (2619 tok/s), M4 Max 128GB (639 tok/s).

GPU compatibility

GPUVRAMQ4 DecodeVerdict
GeForce RTX 409024GB1746 tok/scomfortable
GeForce RTX 509032GB2619 tok/scomfortable
M4 Max 128GB128GB639 tok/scomfortable
M4 Pro 48GB48GB319 tok/scomfortable
M4 Pro 24GB24GB319 tok/scomfortable
A100 PCIe 80 GB80GB3200 tok/scomfortable
H100 SXM5 80 GB80GB6437 tok/scomfortable
GeForce RTX 309024GB1544 tok/scomfortable
Radeon RX 7900 XTX24GB1277 tok/scomfortable
GeForce RTX 408016GB1239 tok/scomfortable
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0