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

Qwen3.5-35B-A3B

by Qwen

36.0B params · image-text-to-text · 1.1k likes · 1.3M downloads

Qwen3.5-35B-A3B is a 36.0B parameter model. At Q4 quantization it requires 18GB of VRAM. It runs comfortably on GeForce RTX 4090 (36 tok/s), GeForce RTX 5090 (54 tok/s), A100 PCIe 80 GB (66 tok/s).

Inference providers

Provider$/1M in$/1M outThroughput
Novita114 tok/s

GPU compatibility

GPUVRAMQ4 DecodeVerdict
GeForce RTX 409024GB36 tok/scomfortable
GeForce RTX 509032GB54 tok/scomfortable
M4 Max 128GB128GB13 tok/stight
M4 Pro 48GB48GB6 tok/stight
M4 Pro 24GB24GB6 tok/stight
A100 PCIe 80 GB80GB66 tok/scomfortable
H100 SXM5 80 GB80GB134 tok/scomfortable
GeForce RTX 309024GB32 tok/scomfortable
Radeon RX 7900 XTX24GB26 tok/stight
M4 Max 64GB64GB13 tok/stight
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0