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

Wizard-Vicuna-30B-Uncensored-GPTQ

by TheBloke

32.5B params · text-generation · 605 likes · 72.8k downloads

Wizard-Vicuna-30B-Uncensored-GPTQ is a 32.5B parameter model. At Q4 quantization it requires 16GB of VRAM. It runs comfortably on GeForce RTX 4090 (40 tok/s), GeForce RTX 5090 (60 tok/s), A100 PCIe 80 GB (73 tok/s).

GPU compatibility

GPUVRAMQ4 DecodeVerdict
GeForce RTX 409024GB40 tok/scomfortable
GeForce RTX 509032GB60 tok/scomfortable
M4 Max 128GB128GB14 tok/stight
M4 Pro 48GB48GB7 tok/stight
M4 Pro 24GB24GB7 tok/stight
A100 PCIe 80 GB80GB73 tok/scomfortable
H100 SXM5 80 GB80GB148 tok/scomfortable
GeForce RTX 309024GB35 tok/scomfortable
Radeon RX 7900 XTX24GB29 tok/stight
M4 Max 64GB64GB14 tok/stight
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0