Does Qwen3.8 Flash Next UD-Q4_K_XL fit in the NVIDIA RTX 3090's memory?

Estimate: the UD-Q4_K_XL file is 111.3 GB and needs about 114.3 GB with an ~8K context; the NVIDIA RTX 3090 has about 23.5 GB usable, so it does not fully fit (part of it runs on the CPU).

Qwen3.8 Flash Next › NVIDIA RTX 3090

Qwen3.8 Flash Next on the NVIDIA RTX 3090

Qwen3.8 Flash Next runs at about 19.0 tokens/s on the NVIDIA RTX 3090 (measured, UD-Q4_K_XL, CUDA, single-stream llama-server decode, part of the model on CPU).

Best measured: 19.0 tokens/s generation (UD-Q4_K_XL, CUDA). The source gives no prompt-processing figure.

For comparison, our size-scaling estimate for this combination is 86.9 tok/s.

Every run

Quant	Backend	Generate tok/s	Prompt tok/s	Test	Build / by / notes	Source
UD-Q4_K_XL	CUDA	19.0	–	decode, llama-server	ZoneOverreach i9-12900K, 128 GB DDR4, --n-cpu-moe 41 (most experts in system RAM).; model file is larger than this card's VRAM, so part of it ran on the CPU/system RAM	unsloth Qwen3.8-Flash-Next GGUF speed thread (Hugging Face)

tg128 = generating 128 tokens; pp512 = reading a 512-token prompt (llama-bench). LocalScore averages several prompt and output lengths, so its numbers run lower than tg128. "decode, llama-server" = single-stream generation speed measured against a running llama-server (no speculative decoding).

Check prices

Amazon Newegg

As an Amazon Associate I earn from qualifying purchases.

Doesn't fit in this card's VRAM

Part of this model ran on the CPU here. A rented 48-80 GB GPU keeps it all in VRAM.

Rent a bigger GPU on RunPod

Affiliate link: if you sign up, RigBench may earn a referral credit. It never affects rankings or numbers.

All results for the NVIDIA RTX 3090 · Qwen3.8 Flash Next on other hardware

Embed this result

HTML

<a href="https://rigbench.pages.dev/model/qwen3-8-flash-next/on/nvidia-rtx-3090/"><img src="https://rigbench.pages.dev/badge/qwen3-8-flash-next--nvidia-rtx-3090.svg" alt="Qwen3.8 Flash Next · NVIDIA RTX 3090: 19.0 tok/s (UD-Q4_K_XL, measured)"></a>

Markdown (GitHub README, forums)

[![Qwen3.8 Flash Next · NVIDIA RTX 3090: 19.0 tok/s (UD-Q4_K_XL, measured)](https://rigbench.pages.dev/badge/qwen3-8-flash-next--nvidia-rtx-3090.svg)](https://rigbench.pages.dev/model/qwen3-8-flash-next/on/nvidia-rtx-3090/)