Rent a GPU

Can the NVIDIA RTX 3090 run Qwen3 Coder Next 80B-A3B?

RTX 3090 × Qwen3 Coder Next 80B-A3B8K context
NoDoes not fit in memory
Memory at Q4_K_M49.1 GB of 24 GB
Weights 48.3 GBKV cache 0.2 GBRuntime 0.6 GB
On the GPU48%of the Q4_K_M weights
In system RAM25.1 GBthe rest spills over
Offload speed~36tokens per second
  1. FP16
  2. Q8_0
  3. Q6_K
  4. Q5_K_M
  5. Q4_K_M
  6. Q3_K_M
  7. Q2_K

No. Even at Q4_K_M, Qwen3 Coder Next 80B-A3B needs about 49.1 GB, but the NVIDIA RTX 3090 has 24 GB usable. You could offload some layers to system RAM, but it will be very slow.

Try other context lengths

Run it on a rented GPU

On a cloud GPU with enough memory, the same command you'd use locally works:

ollama run qwen3-coder-next

At Q4_K_M it needs about 49.1 GB, so pick a GPU with at least 56 GB.

New to this? Step-by-step setup for the RTX 3090

Run it in the cloud

Too big for the NVIDIA RTX 3090? Run Qwen3 Coder Next 80B-A3B on a rented GPU instead.

Rent a cloud GPU by the hour. No upfront cost, stop any time. Not sure whether to rent or buy? Compare the cost.

Rent a GPU on RunPodRent a GPU on Vast.ai

Memory needed by quantization

At 8,192 tokens of context. The black line marks the 24 GB the NVIDIA RTX 3090 can use.

QuantQualityMemoryFits?Speed
FP16Full quality160.2 GBToo big—
Q8_0Near-lossless85.5 GBToo big—
Q6_KExcellent66.2 GBToo big—
Q5_K_MVery good57.5 GBToo big—
Q4_K_MGood, most popular49.1 GBToo big—
Q3_K_MNoticeably degraded39.8 GBToo big—
Q2_KHeavily degraded30.7 GBToo big—

Measured speed from real RTX 3090 owners

Tokens per second that people measured on their own NVIDIA RTX 3090, next to our estimate.

No one has measured Qwen3 Coder Next 80B-A3B on the NVIDIA RTX 3090 yet. The speeds above are estimates. Run the benchmark and yours becomes the first real number on this page.

Have an RTX 3090? Add your real speed

One command times Qwen3 Coder Next 80B-A3B in your own Ollama, shows you exactly what it would send, and asks before sending anything. No sign-up, no personal data.

macOS and Linux

curl -fsSL https://gpufit.dev/bench.sh | sh -s -- qwen3-coder-next

Windows (PowerShell)

$env:GPUFIT_MODEL="qwen3-coder-next"; irm https://gpufit.dev/bench.ps1 | iex

How the benchmark works

GPUs that run Qwen3 Coder Next 80B-A3B comfortably

Check RTX PRO 6000 Blackwell prices

Models that run well on the NVIDIA RTX 3090

Frequently asked questions

Can NVIDIA RTX 3090 run Qwen3 Coder Next 80B-A3B?

No. Even at Q4_K_M, Qwen3 Coder Next 80B-A3B needs about 49.1 GB, but the NVIDIA RTX 3090 has 24 GB usable. You could offload some layers to system RAM, but it will be very slow.

How much VRAM does Qwen3 Coder Next 80B-A3B need?

About 49.1 GB at Q4_K_M with 8,192 tokens of context, and about 160.2 GB at full FP16 precision.

How fast is Qwen3 Coder Next 80B-A3B on the NVIDIA RTX 3090?

It does not fit in the NVIDIA RTX 3090's memory, so most layers would run from system RAM and generation would be very slow.