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

Qwen3.5-9B

by Qwen

9.7B params · image-text-to-text · 680 likes · 1.2M downloads

Qwen3.5-9B is a 9.7B parameter model. At Q4 quantization it requires 5GB of VRAM. It runs comfortably on GeForce RTX 4090 (136 tok/s), GeForce RTX 5090 (203 tok/s), M4 Max 128GB (49 tok/s).

Inference providers

Provider$/1M in$/1M outThroughput
Together AI101 tok/s

GPU compatibility

GPUVRAMQ4 DecodeVerdict
GeForce RTX 409024GB136 tok/scomfortable
GeForce RTX 509032GB203 tok/scomfortable
M4 Max 128GB128GB49 tok/scomfortable
M4 Pro 48GB48GB24 tok/stight
M4 Pro 24GB24GB24 tok/stight
A100 PCIe 80 GB80GB249 tok/scomfortable
H100 SXM5 80 GB80GB501 tok/scomfortable
GeForce RTX 309024GB120 tok/scomfortable
Radeon RX 7900 XTX24GB99 tok/scomfortable
GeForce RTX 408016GB96 tok/scomfortable
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0