Best GPU for Qwen 3.6 in 2026 (35B-A3B MoE Guide)

Qwen 3.6 35B-A3B is a 24GB model, not a 16GB one. VRAM by build, why MoE sizing surprises people, and the GPUs that actually run it.

Quick read: This guide is built to help you match model size, VRAM, and budget before you buy.

The “A3B” in Qwen 3.6 35B-A3B means three billion parameters activate per token, and that is what makes it fast — it generates at roughly the pace of a 3B model while answering like a 35B one. It is also the reason this model gets bought for the wrong hardware. Activation is a compute story, not a memory one. Routing is decided token by token at inference time, so all thirty-five billion parameters have to be resident, and qwen3.6:35b-a3b-q4_K_M is a 24GB download. This is a 24GB-card model. No 16GB card runs it, and neither does the 17GB dense 27B sibling.

Cheapest Card That Fits

NVIDIA GeForce RTX 3090

24GB GDDR6X

24GB used at ~$820 is the least you can spend and still load Qwen 3.6 at Q4. Tight — plan on short context or the 23GB MTP build.

Affiliate links — we may earn a commission at no extra cost to you. Amazon ships globally; Shopee SG covers Singapore & ASEAN.

Qwen 3.6 at a glance

SpecValue
Total parameters35B
Active parameters~3B per token
ArchitectureMixture of Experts (MoE)
Default context262K tokens
Recommended context128K for consumer GPUs
VRAM at Q4_K_M~24GB (~23GB MTP build)
VRAM at Q8_0~39GB
Ollama supportYes (native)

For exact VRAM numbers when running Qwen 3.6 through Ollama (including KV cache overhead), see our Ollama VRAM Requirements guide.

The 262K default context is impressive but aggressive for consumer hardware. On a 24GB card the weights alone consume the card, so context has to stay short; full 262K is a 32GB-plus proposition and the KV cache for it can balloon past 20GB on its own.

VRAM requirements

These are the builds Qwen 3.6 actually ships as, with their sizes taken from the Ollama tag listing rather than estimated from the parameter count:

BuildModel weights+ KV cache (8K)+ KV cache (128K)
27B dense, Q4_K_M~17GB~19GB~23GB
35B-A3B, Q4_K_M (MTP)~23GB~25GB~29GB
35B-A3B, Q4_K_M~24GB~26GB~30GB
27B dense, Q8_0~30GB~32GB~36GB
35B-A3B, Q8_0~39GB~41GB~45GB

The planning decision is simpler than it looks, and it is not the one most guides describe. 24GB is the floor, not the comfortable option. A 3090 or 4090 loads the Q4 weights and has almost nothing left, so you run short context or you run the MTP build for the gigabyte it saves. Long context on this model means 32GB. And if 24GB is your ceiling but you want room to work, the dense 27B at ~17GB is the more sensible model to run on it.

VRAM capacity vs memory bandwidth
RTX 5090 32GB · 1792 GB/s RTX 4090 24GB · 1008 GB/s RX 7900 XTX 24GB · 960 GB/s RTX 3090 (used) 24GB · 936 GB/s RTX 5080 16GB · 960 GB/s RTX 5070 Ti 16GB · 896 GB/s RTX 4070 Ti Super 16GB · 672 GB/s RX 7800 XT 16GB · 624 GB/s RTX 5060 Ti 16GB 16GB · 448 GB/s RTX 4060 Ti 16GB 16GB · 288 GB/s RTX 5070 12GB · 672 GB/s Intel Arc B580 12GB · 456 GB/s RTX 3060 12GB (used) 12GB · 360 GB/s RTX 4060 8GB · 272 GB/s

VRAM capacity memory bandwidth Specs are manufacturer figures. Bar lengths are scaled independently per metric.

Performance benchmarks

Ollama at Q4_K_M with a 128K context window. Throughput is bandwidth-derived rather than benchmarked in a lab (methodology):

GPUtok/sVRAM usedFits?Price
RTX 5090 (32GB)~120 tok/s~24GBYes, with context headroom~$4,900
RTX 4090 (24GB)~85 tok/s~24GBOnly just — short context~$2,200
RTX 3090 (24GB, used)~60 tok/s~24GBOnly just — short context~$820
RTX 5080 (16GB)No~$1,400
RTX 5070 Ti (16GB)No~$1,050
RTX 4070 Ti Super (16GB)No~$800
RTX 4060 Ti 16GBNo~$425
RTX 3060 12GB (used)No~$250

The 16GB cards are not marked “tight” — they are marked “no”. A 24GB file does not load into 16GB at any context length, and there is no smaller official build to fall back on. The next section covers the one way around that.

Throughput is high relative to the model’s size for the reason the name advertises: only ~3B parameters are read per token, so a 35B MoE generates far faster than a dense 35B on the same card. Speed was never the problem with this model. Capacity is.

Check NVIDIA GeForce RTX 4090 on AmazonBuy on Shopee SG

MoE offloading: the 12GB GPU trick

Qwen 3.6 supports --n-cpu-moe in llama.cpp, which offloads inactive expert layers to system RAM. This is the only way to run it below 24GB: the active ~3B parameters stay on the GPU while dormant experts live in CPU memory, so a 16GB or even 12GB card can load a model whose weights are nearly twice its capacity.

It also matters more than it used to, because the alternative is not “run it slower” — it is “do not run it at all”. Budget 32GB of system RAM on top of whatever the card has, and expect a significant drop (roughly 15-25 tok/s on an RTX 3060 12GB with DDR5). Good for evaluating whether the model is worth buying hardware for; not something to build a daily workflow on.

Which GPU should you buy?

GPU Tier List — Local LLM Inference
S
Best Inference
RTX 5090 (32GB)RTX 4090 (24GB)
A
Great for 7B-13B
RTX 4070 Ti Super (16GB)RTX 5080 (16GB)
B
7B Models
RTX 4060 Ti 16GBRTX 3060 12GB
C
Barely Usable
RTX 4060 (8GB)Any 8GB GPU

Cheapest card that fits: RTX 3090 24GB used (~$820). The Q4 weights are 24GB and the card is 24GB, so this works with short context and no room to spare. Take the MTP build for the gigabyte it gives back. Still the best price-per-VRAM on the market.

Best for long context: RTX 5090 (~$4,900). 32GB is the first capacity where 128K context and the Q4 weights coexist without compromise. Expensive, and the only clean single-card answer.

If you already own a 24GB card: run the dense 27B at ~17GB instead. You give up some quality and gain seven gigabytes of working room, which on this model is the difference between usable and constantly out of memory.

16GB cards (5070 Ti, 5080, 4060 Ti 16GB): cannot load either variant. Your options are --n-cpu-moe offloading with the performance cost that carries, or a different model — Qwen 3 14B at ~9.3GB is the sensible 16GB pick.

12GB GPUs: offloading only, and slow enough that it is for evaluation rather than use.

Common mistakes

  • Buying a 16GB card because “A3B” sounds small. The activation count describes compute, not memory. Ollama ships this model as a 24GB file and every byte of it stays resident. This is the most expensive mistake on this page.
  • Comparing Qwen 3.6 to dense 35B models. The 3B active parameters mean it runs 3-4x faster than a dense 35B. Do not use dense model benchmarks to estimate Qwen 3.6 performance.
  • Assuming the newer Qwen has the same hardware needs. Qwen 3.8 is dense rather than MoE and its default build is an 18GB download, so it wants 24GB where this one is comfortable on less — see best GPU for Qwen 3.8 before upgrading on version number alone.
  • Skipping Q4_K_M for Q8. Q8 pushes the model to ~39GB — past every consumer card including the 5090. The quality difference between Q4 and Q8 on MoE models is smaller than on dense models because the active weights are a small fraction of total parameters, so this is a bad trade twice over.

Final verdict

Your goalBest GPUPrice
Cheapest card that loads itRTX 3090 24GB (used)~$820
Long context (128K) without compromiseRTX 5090 32GB~$4,900
Already on 24GB, want working roomrun the dense 27B instead
16GB card, unwilling to offloadpick Qwen 3 14B instead
For Long Context

NVIDIA GeForce RTX 5090

32GB GDDR7

32GB is the first capacity where Qwen 3.6 Q4 and a long context window fit together. The only single-card answer without compromise.

Affiliate links — we may earn a commission at no extra cost to you. Amazon ships globally; Shopee SG covers Singapore & ASEAN.

For the previous generation, see our best GPU for Qwen 3 guide. Running models through Ollama? The best GPU for Ollama article covers multi-model setups. For VRAM planning across all model sizes, how much VRAM for local LLM has the full reference. And if you want to understand quantization trade-offs in depth, read best quantization for local LLM.

Affiliate Disclosure: This article may contain affiliate links. If you purchase through these links, we may earn a commission at no extra cost to you. Learn more
← Back to all guides