unsloth
GLM 5.3 Flash
GLM 5.3 Flash is a 320.8B mixture-of-experts glm5-next model from unsloth, with about 57.0B parameters active per token. Configured context length is 1,048,576 tokens. License: mit. Quant sizes are the GGUF file bytes published on Hugging Face when a file exists.
Parameters321B / 57.0B active
Context1024K
Attentionmla
Licensemit
45 layers · 64 KV heads · head dim 64 · 34 linear-attention layers · GGUF repo · config
Quant file sizes
| Quant | File size | Source |
|---|---|---|
| BF16 | 597.6 GB | Hugging Face file bytes |
| Q8_0 | 317.6 GB | Hugging Face file bytes |
| UD-Q6_K_XL | 271.8 GB | Hugging Face file bytes |
| UD-Q5_K_XL | 223.8 GB | Hugging Face file bytes |
| UD-Q4_K_XL | 186.0 GB | Hugging Face file bytes |
| UD-IQ4_XS | 146.1 GB | Hugging Face file bytes |
| UD-Q3_K_XL | 137.4 GB | Hugging Face file bytes |
| UD-IQ3_XXS | 112.1 GB | Hugging Face file bytes |
| UD-Q2_K_XL | 101.3 GB | Hugging Face file bytes |
| UD-IQ2_XXS | 94.9 GB | Hugging Face file bytes |
| UD-IQ1_M | 90.9 GB | Hugging Face file bytes |
| UD-IQ1_S | 86.7 GB | Hugging Face file bytes |
On common machines
| Device | Verdict | Quant | Speed |
|---|---|---|---|
| GeForce RTX 4060 Ti 8GB | Won't fit | UD-IQ1_S | — |
| GeForce RTX 3060 12GB | Won't fit | UD-IQ1_S | — |
| GeForce RTX 4060 Ti 16GB | Won't fit | UD-IQ1_S | — |
| GeForce RTX 4090 | Won't fit | UD-IQ1_S | — |
| GeForce RTX 5090 | Won't fit | UD-IQ1_S | — |
| Apple M4 Pro | Won't fit | UD-IQ1_S | — |
| Apple M5 | Won't fit | UD-IQ1_S | — |
| CPU only · DDR5-5600 | Won't fit | UD-IQ1_S | — |
Smallest devices that fit it fully
- NVIDIA H200 SXM — Runs, UD-IQ3_XXS, 123–168 tok/s
- NVIDIA B200 — Runs, UD-IQ4_XS, 157–215 tok/s
Run it
Commands for the GeForce RTX 4060 Ti 16GB estimate (UD-IQ1_S).
llama.cpp
llama-server -hf unsloth/GLM-5.3-Flash-GGUF:UD-IQ1_S -c 2048 -ngl 0
Ollama
ollama run glm-5.3-flash
LM Studio
lms get unsloth/GLM-5.3-Flash-GGUF # Then load quant UD-IQ1_S