Can 3x DGX Spark (GB10) run GLM-5.3?

Probably only partly: the weights are larger than its memory, so part of the model likely ran from system RAM. Measured 31.3 tok/s generation with part of the model offloaded to system RAM (TensorFold, 3 nodes, 1 stream, EXL3 2.75-bit + DSpark; source).

Every measured result for GLM-5.3 on 3x DGX Spark (GB10)

Build	Quant / setup	Engine	tok/s
3x DGX Spark: GLM-5.3 EXL3 2.75-bit on TensorFold, 31.3 tok/s (Mia's AI Lab recipe)	EXL3 2.75-bit + DSpark	TensorFold, 3 nodes, 1 stream	31.3 CPU offload	source

Memory: 3x DGX Spark (GB10) has 384 GB. GLM-5.3 weights are about 460.7 GB at Q4_K_M (753B params × 4.89 bits ÷ 8, before context/KV cache). Arithmetic, not a measurement; long contexts need more. Check other quants in the VRAM calculator.

Speeds are decode (generation) tokens/sec as reported by each source; quant, engine and context differ, so small gaps are noise.

All measured speeds for GLM-5.3: best home rig for GLM-5.3.

Other models measured on 3x DGX Spark (GB10): GLM-5.3 Flash

Not enough VRAM? Rent before you buy. Run GLM-5.3 on a bigger GPU on a rented 24–96 GB cloud GPU by the hour and see the real speed before spending €1,000+. Try it on Runpod
Referral link: if you sign up with Google and add $10, Runpod gives you and RigBase bonus credit (terms). It never affects our data.