Alibaba Cloud27B~18 GB VRAM at Q4_K_M

Qwen3.8 27B — VRAM, speed & local setup

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

Written by Jakub Rusinowski · Last updated September 6, 2026

27.78B dense, natively multimodal — text, images and video in. 262,144-token native context, extendable to 1M with YaRN, Apache 2.0. 17.6 GB at Q4_K_M puts it inside a 24 GB card with room for context, which is what makes it the current default recommendation at the single-GPU tier.

Qwen3.8 27B needs about 18 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.

Logic90

Creative86

Coding91

VRAM and speed by quantization

Quoted against NVIDIA RTX 4090 (24 GB). Includes the KV cache at 8K context, so it reads higher than the headline figure.

Quant	VRAM	Speed (est.)	Fit
Q2_K 2.63 bpw	10.6 GB	~67 tok/s	Fits
Q3_K_M 3.41 bpw	13.3 GB	~54 tok/s	Fits
Q4_K_M 4.83 bpw	18.3 GB	~40 tok/s	Fits
Q5_K_M 5.67 bpw	21.2 GB	~35 tok/s	Fits
Q6_K 6.56 bpw	24.3 GB	~5 tok/s	Offload
Q8_0 8.50 bpw	31 GB	~4 tok/s	Offload
F16 16.00 bpw	57.1 GB	—	Too big

Black marker = usable memory on the NVIDIA RTX 4090 (24 GB). Estimates from the memory-bandwidth roofline on the methodology page. Qwen3.8 27B VRAM calculator →

Get Qwen3.8 27B running

The cheapest catalogued GPU that runs Qwen3.8 27B is the AMD Radeon RX 7900 XT (20 GB).

Buy this hardware AMD Radeon RX 7900 XT 20GB — 20 GB VRAM · 315 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.

AMD Radeon RX 7900 XT 20GB

20 GB VRAM · 315 W board power

2026 prices are volatile — check the current listing.

Check price on Amazon

How to run Qwen3.8 27B

Install Ollama, then run:

$ollama run qwen3.8:27b

Weights on Hugging Face: Qwen/Qwen3.8-27B ↗

Specifications

Verified — Checked against the primary source — the model card or the vendor spec page — and corroborated by a second independent source.

Parameters: 27.78 Billion
Context window: 262,144
Architecture: Dense Transformer (native multimodal)
Provider: Alibaba Cloud
Licence: Apache 2.0
Practical deployment: Single consumer GPU
Specified at: Q4_K_M
System RAM: 32 GB
Record updated: 2026-09-06

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 forcodinggeneral purposemultimodallong context

Can I run Qwen3.8 27B on my GPU?

Other Qwen3.8 sizes

Qwen3.8 27B — frequently asked questions

How much VRAM does Qwen3.8 27B need?

About 18 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 Qwen3.8 27B run on an RTX 4090 (24 GB)?

Yes. Qwen3.8 27B needs about 18 GB at Q4_K_M, inside a 24 GB card, at an estimated 40 tokens/sec.

How do I run Qwen3.8 27B locally?

Install Ollama and run `ollama run qwen3.8:27b`. That pulls the weights and starts a local OpenAI-compatible endpoint; after the download nothing leaves the machine.

What other sizes does Qwen3.8 come in?

Qwen3.8 27B (18 GB), Qwen3.8-Max (1450 GB), Qwen3.8-Flash-Next (109 GB). Every size shares the family's training and licence; the larger ones score higher and need proportionally more memory.