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

Qwen3-Coder-30B-A3B-Instruct-GGUF

by unsloth

30.0B params · text-generation · 529 likes · 178.8k downloads

Qwen3-Coder-30B-A3B-Instruct-GGUF is a 30.0B parameter model. At Q4 quantization it requires 15GB of VRAM. It runs comfortably on GeForce RTX 4090 (43 tok/s), GeForce RTX 5090 (65 tok/s), A100 PCIe 80 GB (80 tok/s).

GPU compatibility

GPUVRAMQ4 DecodeVerdict
GeForce RTX 409024GB43 tok/scomfortable
GeForce RTX 509032GB65 tok/scomfortable
M4 Max 128GB128GB16 tok/stight
M4 Pro 48GB48GB8 tok/stight
M4 Pro 24GB24GB8 tok/stight
A100 PCIe 80 GB80GB80 tok/scomfortable
H100 SXM5 80 GB80GB161 tok/scomfortable
GeForce RTX 309024GB38 tok/scomfortable
Radeon RX 7900 XTX24GB32 tok/scomfortable
M4 Max 64GB64GB16 tok/stight
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0