Can NVIDIA DGX Spark (GB10) · 128 GB run GLM-5.3 (753B-A40B)?
128 GB unified · 273 GB/s memory bandwidth · GLM-5.3 (753B-A40B) has 753B parameters (40B active per token).
Short answer
No. It doesn't fit in memory.
Needs ≈348 GB but only ≈120 GB is available after the OS, other apps and Open WebUI.
Why it doesn’t fit
Even its smallest practical version (Q3_K_M) needs about 348 GB for the model, its context memory and the runtime. After the operating system (≈4.0 GB), a browser with a few tabs (≈3.0 GB) and Open WebUI (≈1.0 GB), NVIDIA DGX Spark (GB10) · 128 GB has about 120 GB left, so it’s 228 GB short.
Ways to make it work
Close everything else
Not enough on its own: with every other app closed it’s still about 225 GB short.
Pick a smaller model from the same family
Get more memory
You’d want at least 356 GB of memory in total. The machines below are the cheapest that run it comfortably.
Check your exact setup
These pages assume typical settings. A shorter context or a lighter chat app can change the answer.
Check NVIDIA DGX Spark (GB10) · 128 GB
The same applies to every workload we rate (chat, coding questions, coding in a repository, agentic coding and long documents): they all need at least as much memory as chat.
Other models for NVIDIA DGX Spark (GB10) · 128 GB
The best-rated models for chat on this machine.
- gpt-oss-120b≈55 tok/sExcellent
- Qwen3.6 35B-A3B60–80 tok/sExcellent
- Qwen3-Next 80B-A3B Instruct60–80 tok/sExcellent
- Gemma 4 26B-A4B45–60 tok/sExcellent
- Qwen3.5 35B-A3B60–80 tok/sExcellent
- LFM2.5 24B-A2B120–160 tok/sExcellent
Cheapest machines for GLM-5.3 (753B-A40B)
Hardware rated comfortable or better for chat, lowest price first.
- Mac Studio M5 Ultra 512 GB$7,099 · 16–25 tok/sComfortable
- Mac Studio M3 Ultra 512 GB$7,099 · 11–15 tok/sComfortable
Same chip, other memory: 64 GB
Different tool, context size or workload?
These pages use typical settings. The full check models your exact setup.