Does GLM-5.3-Flash UD-Q2_K_XL fit in the NVIDIA DGX Spark (GB10)'s memory?

Estimate: the UD-Q2_K_XL file is 108.7 GB and needs about 111.7 GB with an ~8K context; the NVIDIA DGX Spark (GB10) has about 96.0 GB usable, so it does not fully fit (part of it runs on the CPU).

Home › Models › GLM-5.3-Flash › NVIDIA DGX Spark (GB10)

GLM-5.3-Flash on the NVIDIA DGX Spark (GB10)

GLM-5.3-Flash runs at about 17.5 tokens/s on the NVIDIA DGX Spark (GB10) (measured, UD-Q2_K_XL, CUDA, single-stream llama-server decode).

Best measured generation17.5tok/s

UD-Q2_K_XL · CUDA

The source gives no prompt-processing figure.

For comparison, our size-scaling estimate for this combination is 24.2 tok/s.

Every run

Quant	Backend	Generate tok/s	Prompt tok/s	Test	Build / by / notes	Source
UD-Q2_K_XL	CUDA	17.5	–	decode, llama-server, no MTP	Single DGX Spark, 32K context, temp 0, warmed; the no-MTP baseline (MTP figures on the same page are not used). llama.cpp from upstream PR #27752.	Weschera/glm53-flash-one-spark (GitHub)

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.

No card yet?

Try it on a rented GPU (RunPod)

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

All results for the NVIDIA DGX Spark (GB10) · GLM-5.3-Flash on other hardware

Embed this result

HTML

<a href="https://rigbench.pages.dev/model/glm-5-3-flash/on/nvidia-dgx-spark-gb10/"><img src="https://rigbench.pages.dev/badge/glm-5-3-flash--nvidia-dgx-spark-gb10.svg" alt="GLM-5.3-Flash · NVIDIA DGX Spark (GB10): 17.5 tok/s (UD-Q2_K_XL, measured)"></a>

Markdown (GitHub README, forums)

[![GLM-5.3-Flash · NVIDIA DGX Spark (GB10): 17.5 tok/s (UD-Q2_K_XL, measured)](https://rigbench.pages.dev/badge/glm-5-3-flash--nvidia-dgx-spark-gb10.svg)](https://rigbench.pages.dev/model/glm-5-3-flash/on/nvidia-dgx-spark-gb10/)