Skip to content
localmodel.run

Text model · GLM

GL GLM-4.7-Flash: RAM and VRAM requirements

GLM-4.7-Flash needs about 19.2 GB to run at Q4_K_M (the Q4_K_M GGUF file is ~17.05 GB to download; KV cache and overhead add the rest), or about 31.8 GB at Q8_0. The lightest hardware that runs it is Nvidia GeForce RTX 4090 (24GB).

GLM family · 30B params (Mixture-of-Experts: activates only 3B of 30B params per token, so generation is faster than the total size suggests) · released Jan 2026.

Shopping for hardware? See what runs GLM-4.7-Flash →

License MIT · Commercial OK ↓ 2.1M/mo ♥ 1.8K on HuggingFace
Q4_K_M GGUF
17.05 GB
Q8_0 GGUF
29.66 GB
Memory @ Q4 (4k)
~19.2 GB
Context
200 k

Q4_K_M and Q8_0 are GGUF download (file) sizes. Memory @ Q4 includes the KV cache and overhead, so it runs larger than the file on disk.

Will it run on your device?

GLM-4.7-Flash runs on 13 of 40 tracked devices at Q4_K_M.

12 run well 1 tight fit 27 too small
Fit check Q4_K_M
No, not enough memory
needs 19.2 GB usable 10.5 GB
See the full breakdown

Memory breakdown

Weights (Q4_K_M)17.05 GB
+
KV cache (4k)1.3 GB
+
Overhead0.8 GB
=
Total19.2 GB

How context length changes it

4k context ~19.2 GB 32k context ~28.4 GB 128k context ~60 GB

Longer context grows the KV cache, which for GLM-4.7-Flash is sized by its 3B active params, not the full 30B. It needs ~19.2 GB at 4k and ~60 GB at 128k.

Quantization sizes

GGUF quantson disk
Quantization Size on disk
Q2_K 12.6 GB est
Q3_K_M 14.7 GB est
Q4_K_M (default) 17.05 GB
Q5_K_M 21.4 GB est
Q6_K 24.6 GB est
Q8_0 29.66 GB
FP16 55.79 GB

Lower quant = smaller and faster, slightly lower quality. Q4_K_M is the common default.

Run it

Ollama
$ ollama run glm-4.7-flash:latest
llama.cpp
$ llama-cli -hf unsloth/GLM-4.7-Flash-GGUF:Q4_K_M
LM Studio
$ lms get unsloth/GLM-4.7-Flash-GGUF

Which devices can run GLM-4.7-Flash?

Apple Silicon Macs

RAM-only laptops

iPhone & iPad

Android

NVIDIA GPUs

AMD GPUs

Same job, different size

Pick by what fits your memory: step down to free up VRAM, or step up if you have headroom.

Next size down Granite 4.1 30B 28.9B Next size up Qwen3 30B-A3B 30.5B

Similar models

Head-to-head

FAQ

How much VRAM or RAM does GLM-4.7-Flash need?

At Q4_K_M, GLM-4.7-Flash needs about 19.2 GB (weights ~17.05 GB + KV cache + overhead) at a 4k context. At Q8_0 budget ~31.8 GB.

What is the Q4_K_M GGUF file size of GLM-4.7-Flash?

The Q4_K_M GGUF file is about 17.05 GB to download, and the Q8_0 GGUF is about 29.66 GB. That is the weights file on disk; to run it you also need room for the KV cache and overhead, so budget ~19.2 GB of memory at Q4_K_M.

Can GLM-4.7-Flash run on a laptop?

GLM-4.7-Flash is large; you need a 24 GB+ GPU or a 32-48 GB Mac at Q4_K_M.

Is GLM-4.7-Flash cheaper to run because it is a MoE model?

It is faster, not lighter. GLM-4.7-Flash activates only 3B of 30B params per token (so it runs quickly), but all experts must stay in memory, so it still needs memory for the full 30B.

Can I use GLM-4.7-Flash commercially?

Yes. GLM-4.7-Flash is licensed MIT, which permits commercial use.

Understand the numbers

Short guides to the ideas behind GLM-4.7-Flash's memory and quant figures.

MoE: 30B total / 3B active (64 routed experts, 4 activated per token). Z.ai calls it the strongest model in the 30B class. ~200K context (config max_position_embeddings 202752). Sizes from unsloth's GGUF repo. MIT.

Sources

Last validated 2026-08-03. Memory figures are estimates. See methodology.