Alibaba Cloud35B (3B active)~22 GB VRAM at Q4_K_M

Qwen 3.5 35B-A3B — VRAM, speed & local setup

Name: LLM Configurator — GPU VRAM Checker
Author: LLM Configurator

Written by Jakub Rusinowski · Last updated February 24, 2026

MoE variant with ~36B total parameters but only 3B active per token — best MoE efficiency in the family. Fits on an RTX 4090/3090 at Q4 (~21.4 GB) while outperforming most 70B dense models. Note: third-party GGUF imports can fail in Ollama due to a separate mmproj vision file; use the official `qwen3.5:35b` library tag, which bundles vision support.

Qwen 3.5 35B-A3B needs about 22 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache. On Apple Silicon that figure comes out of unified memory.

Logic91

Creative89

Coding92

VRAM and speed by quantization

Quoted against NVIDIA RTX 4090 (24 GB). Weights plus overhead only: this model's architecture is unpublished, so no KV cache is added.

Quant	VRAM	Speed (est.)	Fit
Q2_K 2.63 bpw	12.3 GB	~209 tok/s	Fits
Q3_K_M 3.41 bpw	15.7 GB	~193 tok/s	Fits
Q4_K_M 4.83 bpw	21.9 GB	~170 tok/s	Tight
Q5_K_M 5.67 bpw	25.6 GB	~24 tok/s	Offload
Q6_K 6.56 bpw	29.5 GB	~22 tok/s	Offload
Q8_0 8.50 bpw	38 GB	~19 tok/s	Offload
F16 16.00 bpw	70.8 GB	—	Too big

Black marker = usable memory on the NVIDIA RTX 4090 (24 GB). Estimates from the memory-bandwidth roofline on the methodology page. Qwen 3.5 35B-A3B VRAM calculator →

Get Qwen 3.5 35B-A3B running

The cheapest catalogued GPU that runs Qwen 3.5 35B-A3B is the NVIDIA GeForce RTX 5090 (32 GB).

Buy this hardware NVIDIA GeForce RTX 5090 32GB — 32 GB VRAM · 575 W board power Deploy in the cloud now RTX 4090 on RunPod — from $0.34/hr · rate checked 2026-07

or compare on Vast.ai from $0.35/hr (typical low · varies)

As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.

Affiliate disclosure: Some links on this page are affiliate links — if you buy through them, LLM Configurator may earn a commission at no extra cost to you. As an Amazon Associate, LLM Configurator earns from qualifying purchases.

NVIDIA GeForce RTX 5090 32GB

32 GB VRAM · 575 W board power

2026 prices are volatile — check the current listing.

Check price on Amazon

How to run Qwen 3.5 35B-A3B

Install Ollama, then run:

$ollama run qwen3.5:35b-a3b

Weights on Hugging Face: Qwen/Qwen3.5-35B-A3B-Instruct ↗

Specifications

Curated — A hand-written entry from before this catalogue recorded its sources. The figures are long-standing but their provenance is not on file.

Parameters: ~36 Billion (3B active)
Context window: 262,144
Architecture: Hybrid Gated DeltaNet + MoE
Provider: Alibaba Cloud
Licence: Apache 2.0
Practical deployment: Single consumer GPU
Specified at: Q4_K_M
System RAM: 32 GB
Record updated: 2026-02-24

LicenceApache-2.0Commercial use permitted

Commercial use permitted. No usage restrictions beyond attribution.

Quality and use cases

Scores as published by the model’s authors or an independent evaluator — quality, not throughput, and not measured by us.

Best forreasoningcodinggeneral useconsumer gpu

Can I run Qwen 3.5 35B-A3B on my GPU?

Other Qwen 3.5 sizes

Qwen 3.5 35B-A3B — frequently asked questions

How much VRAM does Qwen 3.5 35B-A3B need?

About 22 GB at Q4_K_M — quantized weights plus framework overhead, before any KV cache. The cache grows with context length and is added on top; the table above folds it in. Apple Silicon counts unified memory toward the same figure.

Does Qwen 3.5 35B-A3B run on an RTX 4090 (24 GB)?

Yes. Qwen 3.5 35B-A3B needs about 22 GB at Q4_K_M, inside a 24 GB card, at an estimated 170 tokens/sec.

How do I run Qwen 3.5 35B-A3B locally?

Install Ollama and run `ollama run qwen3.5:35b-a3b`. That pulls the weights and starts a local OpenAI-compatible endpoint; after the download nothing leaves the machine.

What other sizes does Qwen 3.5 come in?

Qwen 3.5 0.8B (1 GB), Qwen 3.5 2B (2 GB), Qwen 3.5 4B (3 GB), Qwen 3.5 9B (6 GB), Qwen 3.5 27B (17 GB), Qwen 3.5 35B-A3B (22 GB), Qwen 3.5 122B-A10B (74 GB), Qwen 3.5 397B-A17B (240 GB). Every size shares the family's training and licence; the larger ones score higher and need proportionally more memory.