No — not on this device

GLM-5.3 744B-A40B at Q4_K_M needs 424.7 GB against 96.0 GB usable, and the shortfall of 328.7 GB is more than 16 GB of system RAM can cover at a tolerable speed. A smaller sibling or a lower quantisation is the honest answer here.

Does not fit 8K context Q4_K_M · 423.4 GB GLM-5.3 License Released 25 Aug 2026 Not in the Ollama library

Same base as GLM-5.2, new post-training: Z.ai's most capable open-weights coder (Terminal-Bench 3.0 28.3, DeepSWE 66.9). Not MIT any more — its own GLM-5.3 licence. Same ceiling as 5.2: 512 GB of unified memory at Q4.

Hardware requirements What hardware do I need to run GLM-5.3 744B-A40B? →

No single GPU or Mac in the catalogue holds it at Q4_K_M — the page says what comes closest. Minimum, recommended, high-quality and long-context tiers, with all 114 devices checked at every quantisation.

Fits instead: GLM-4.7-Flash 30B-A3B (18.6 GB)

The VRAM budget

weights 423.4 GB
Weights 423.4 GB KV cache @ 8K 0.69 GB Runtime overhead 0.6 GB Over budget 328.7 GB past 96.0 GB

Quantisation ladder

QuantWeightsTotal @ 8KMax contextTok/sQualityFit
Q8_0 745.1 GB 746.4 GB — ~1.8 −0.1% ppl 650.4 GB over
Q6_K 575.1 GB 576.3 GB — ~2.3 −0.4% ppl 480.3 GB over
Q5_K_M 497.0 GB 498.3 GB — ~2.7 −0.8% ppl 402.3 GB over
Q4_K_M 423.4 GB 424.7 GB — ~3.2 −1.9% ppl 328.7 GB over
Q3_K_M 342.8 GB 344.0 GB — ~3.9 −5.4% ppl 248.0 GB over
Q2_K 293.7 GB 294.9 GB — ~4.6 −15% ppl 198.9 GB over

Quality is the published perplexity delta against f16 weights. Max context assumes an f16 KV cache; q8_0 roughly doubles it. This model uses multi-head latent attention, so its cache is a compressed latent rather than full K and V.

How to run it

terminal
$ llama-server \
    -hf zai-org/GLM-5.3:Q4_K_M \
    -c 8192 -ngl 17

The engine underneath most of the others. Every knob is exposed. More on llama.cpp.

01Download is 423.4 GB. Keep it on an SSD — a first load off a spinning disk takes minutes.
02The GPU shares one pool with the operating system, and these numbers count about 96.0 GB of 128 GB as its share. Under Linux the GPU can usually address more than that, so treat the verdict as conservative.
03If it falls back to CPU silently, drop the context first, then step down a quantisation.
See all models for this rig Compare with another model

Related models on a DGX Spark · 128 GB

ModelTotalTok/s est.
GLM-5.2 744B-A40B
753B MoE · same family
Q4_K_M · 424.7 GB — Won't fit
GLM-5.3-Flash 320B-A18B
321B MoE · same family
Q4_K_M · 181.2 GB — Won't fit
GLM-4.7-Flash 30B-A3B
31.2B MoE · same family
Q4_K_M · 18.6 GB 42 Runs great
DeepSeek-R1 671B
671B MoE · similar size
Q4_K_M · 378.4 GB — Won't fit
Kimi K2.6 1T-A32B
1027B MoE · similar size
Q4_K_M · 578.6 GB — Won't fit
DeepSeek V4.1 Flash 552B-A16B
552B MoE · similar size
Q4_K_M · 311.0 GB — Won't fit
Qwen3.5 397B-A17B
403B MoE · similar size
Q4_K_M · 227.4 GB — Won't fit

Every model against the DGX Spark · 128 GB →

GLM-5.3 744B-A40B on other hardware

DeviceTotalTok/s est.
GeForce RTX 4060 Laptop
7.0 GB usable · 256 GB/s
424.7 GB — Won't fit
GeForce RTX 3060 12 GB
10.6 GB usable · 360 GB/s
424.7 GB — Won't fit
M4 · 16 GB
10.7 GB usable · 120 GB/s
424.7 GB — Won't fit
GeForce RTX 4090
22.4 GB usable · 1008 GB/s
424.7 GB — Won't fit
Radeon RX 7900 XTX
22.4 GB usable · 960 GB/s
424.7 GB — Won't fit
M3 Max · 64 GB
48.0 GB usable · 400 GB/s
424.7 GB — Won't fit

All 114 devices, by tier →