zai-org
GLM-4.7-Flash
- Parameters
- 29.9B
- Max context length
- 202,752
- Architecture
- deepseek2
- License
- Commercial OK
- Base model
- zai-org/GLM-4.7-Flash
- License
- mit
About GLM-4.7-Flash
GLM-4.7-Flash is a GGUF model based on zai-org/GLM-4.7-Flash on Hugging Face. GGUF files are published by 4 sources (unsloth/GLM-4.7-Flash-GGUF, lmstudio-community/GLM-4.7-Flash-GGUF, ggml-org/GLM-4.7-Flash-GGUF and others). It has about 29.9B parameters, the deepseek2 architecture, a maximum context length of 202,752 tokens (values written in the GGUF).
There are 41 quantizations, with file sizes from 7.62 GB (IQ2_XXS) to 59.9 GB (BF16) (actual sizes on Hugging Face).
Running Q4_K_M (18.3 GB) with an 8,192-token context needs an estimated 19.6 GB of memory, and an estimated 20.9 GB at 32,768 tokens. For 8,192 tokens, a GPU with 24 GB or more of VRAM (such as GeForce RTX 3090 Ti, GeForce RTX 4090), or a Mac with 24 GB or more of unified memory, is a rough guide (macOS caps how much memory the GPU can use, so leave some headroom). The smallest, IQ2_XXS (7.62 GB), needs an estimated 8.92 GB at 8,192 tokens.
The original model is licensed under MIT License, which allows commercial use (keep the license text and copyright notice when redistributing). Check the original license for the terms of use.
File size and estimated memory per quantization
- Up to 8 GB
- Up to 16 GB
- Up to 24 GB
- Up to 48 GB
- Up to 128 GB
- More
File sizes are the actual values on Hugging Face. Memory is estimated for each context length (KV cache f16). The formula is in How memory is calculated.
All memory figures are Estimated (KV cache f16). It should run on a setup with at least this much VRAM (or Mac unified memory). Click a value to see the devices and models that fit that amount (Find by GPU / Mac).
4 sources publish GGUF files of this model. Quantization types and file sizes differ slightly between sources, so there is a table for each source below.
About the "quality guide" (how far a quantization drifts from the original)
For each quantization type, how far the output drifts from the original model, grouped into six levels from values llama.cpp measured on Llama-3-8B.
The value is mean KLD (the difference between the original and quantized model in next-token probabilities); smaller means closer to the original. Quantizations made with an imatrix (importance data) come out slightly smaller. Levels use the value without an imatrix (or with one, for types that only have that).
These values were not measured on this model. Larger models and MoE models tend to drift less with the same type. Distributor-specific mixed types (such as UD-Q4_K_XL) cannot be matched by type name alone, so they show "-".
| Quantization | Quality guide | KLD (no imatrix) | KLD (imatrix) | Size (Llama-3-8B) |
|---|---|---|---|---|
| Q8_0 | Nearly identical | 0.0014 | - | 7.96 GiB |
| Q6_K | Very small | 0.0055 | - | 6.14 GiB |
| Q5_K_M | Very small | 0.0108 | - | 5.33 GiB |
| Q5_K_S | Small | 0.0166 | - | 5.21 GiB |
| Q5_1 | Small | 0.0180 | - | 5.65 GiB |
| Q5_0 | Small | 0.0222 | - | 5.21 GiB |
| Q4_K_M | Small | 0.0313 | 0.0282 | 4.58 GiB |
| Q4_K_S | Medium | 0.0431 | 0.0320 | 4.37 GiB |
| IQ4_NL | Small | - | 0.0357 | 4.35 GiB |
| IQ4_XS | Small | - | 0.0363 | 4.14 GiB |
| Q4_1 | Medium | 0.0717 | - | 4.78 GiB |
| Q4_0 | Medium | 0.0719 | - | 4.34 GiB |
| Q3_K_L | Medium | 0.0872 | 0.0731 | 4.03 GiB |
| Q3_K_M | Medium | 0.1019 | 0.0844 | 3.74 GiB |
| IQ3_M | Medium | - | 0.1025 | 3.53 GiB |
| IQ3_S | Medium | - | 0.1113 | 3.42 GiB |
| IQ3_XS | Large | - | 0.1387 | 3.28 GiB |
| IQ3_XXS | Large | - | 0.1836 | 3.05 GiB |
| Q3_K_S | Large | 0.2282 | 0.1998 | 3.41 GiB |
| IQ2_M | Large | - | 0.3260 | 2.74 GiB |
| Q2_K | Very large | 0.4451 | 0.3322 | 2.96 GiB |
| Q2_K_S | Very large | - | 0.4034 | 2.96 GiB |
| IQ2_S | Very large | - | 0.4392 | 2.56 GiB |
| IQ2_XS | Very large | - | 0.5463 | 2.43 GiB |
| IQ2_XXS | Very large | - | 0.8120 | 2.24 GiB |
| IQ1_M | Very large | - | 1.3931 | 2.01 GiB |
| IQ1_S | Very large | - | 2.2546 | 1.88 GiB |
Source: llama.cpp (MIT), measured on Llama-3-8B against FP16, Wikitext-2, revision f364eb6f. Original table · Reading and choosing quantizations
Last checked: Nov 11, 2026
- Architecture
- deepseek2
- Supported before llama.cpp b4409
- Parameters
- 29.9B
- Max context length
- 202,752
- License
- mit Commercial OK
Ollama ollama run hf.co/unsloth/GLM-4.7-Flash-GGUF:Q4_K_M
llama.cpp llama-server -hf unsloth/GLM-4.7-Flash-GGUF:Q4_K_M
GGUF source
lmstudio-community/GLM-4.7-Flash-GGUF
Converted and published by a third party About this label
Last checked: Nov 1, 2026
- Architecture
- deepseek2
- Supported before llama.cpp b4409
- Parameters
- 29.9B
- Max context length
- 202,752
- License
- mit Commercial OK
| Quantization | File size Measured | Quality guide | 4,096 | 8,192 | 16,384 | 32,768 | 65,536 | 131,072 | 202,752 max |
|---|---|---|---|---|---|---|---|---|---|
| Q4_K_M | 18.1 GB | Small | 19.2 GB | 19.4 GB | 19.9 GB | 20.8 GB | 22.5 GB | 26.1 GB | 30.0 GB |
| Q6_K | 24.6 GB | Very small | 25.7 GB | 25.9 GB | 26.4 GB | 27.2 GB | 29.0 GB | 32.6 GB | 36.4 GB |
| Q8_0 | 31.8 GB | Nearly identical | 32.9 GB | 33.1 GB | 33.6 GB | 34.5 GB | 36.2 GB | 39.8 GB | 43.7 GB |
Ollama ollama run hf.co/lmstudio-community/GLM-4.7-Flash-GGUF:Q4_K_M
llama.cpp llama-server -hf lmstudio-community/GLM-4.7-Flash-GGUF:Q4_K_M
Last checked: Oct 24, 2026
- Architecture
- deepseek2
- Supported before llama.cpp b4409
- Parameters
- 29.9B
- Max context length
- 202,752
- License
- mit Commercial OK
| Quantization | File size Measured | Quality guide | 4,096 | 8,192 | 16,384 | 32,768 | 65,536 | 131,072 | 202,752 max |
|---|---|---|---|---|---|---|---|---|---|
| Q4_K | 18.2 GB | - | 19.3 GB | 19.5 GB | 20.0 GB | 20.9 GB | 22.6 GB | 26.2 GB | 30.1 GB |
| Q8_0 | 31.8 GB | Nearly identical | 32.9 GB | 33.1 GB | 33.6 GB | 34.5 GB | 36.2 GB | 39.8 GB | 43.7 GB |
Ollama ollama run hf.co/ggml-org/GLM-4.7-Flash-GGUF:Q4_K
llama.cpp llama-server -hf ggml-org/GLM-4.7-Flash-GGUF:Q4_K
GGUF source
bartowski/zai-org_GLM-4.7-Flash-GGUF
Converted and published by a third party About this label
Last checked: Oct 24, 2026
- Architecture
- deepseek2
- Supported before llama.cpp b4409
- Parameters
- 29.9B
- Max context length
- 202,752
- License
- mit Commercial OK
Ollama ollama run hf.co/bartowski/zai-org_GLM-4.7-Flash-GGUF:Q4_K_M
llama.cpp llama-server -hf bartowski/zai-org_GLM-4.7-Flash-GGUF:Q4_K_M
Third-party evaluations
Values from other organizations for the original model before quantization. Not this site's own rating. Values are for the original model before quantization. GGUF quantization (especially small quantizations) can lower performance.
Human preference (LMArena)
1,364 1,359–1,370
glm-4.7-flash · 11,937 votes · Values as of Nov 9, 2026
Source: LMArena (CC BY 4.0). Original leaderboard · See in ranking
License and commercial use
MIT License Commercial OK
Allows commercial use, modification and redistribution. When redistributing, keep the copyright notice and license text.
Based on the license on the original model's Hugging Face page. Original license text (checked Oct 18, 2026)
The summary states facts this site read from the original text and is not legal advice. Always check the original license and its latest terms before use. All licenses
How to run
Example commands using Q4_K_M. To use another quantization, replace the name with one from the table above. Commands for files from other sources are in each source's section. How to run local LLMs
Ollama
Run it by pointing directly at the GGUF on Hugging Face.
ollama run hf.co/unsloth/GLM-4.7-Flash-GGUF:Q4_K_M
llama.cpp
Pass the repository and quantization name to llama-server (or llama-cli) with -hf.
llama-server -hf unsloth/GLM-4.7-Flash-GGUF:Q4_K_M
LM Studio
Paste unsloth/GLM-4.7-Flash-GGUF into the search box, download it, and choose the quantization from the file list.
Other models from zai-org
-
320.8B
zai-org
GLM-5.3-Flash
- UD-IQ1_S size
- 93.1 GB
- Memory at 8K
- 94.2 GB
Measured Estimated VRAM 96 GB+
-
753.9B
zai-org
GLM-5.3
- UD-IQ1_S size
- 217 GB
- Memory at 8K
- 218 GB
Measured Estimated
-
110.5B
zai-org
GLM-4.5-Air
- Q4_K_M size
- 73.0 GB
- Memory at 8K
- 75.6 GB
Measured Estimated VRAM 96 GB+
-
0.9B
zai-org
GLM-OCR
- Q8_0 size
- 0.95 GB
- Memory at 8K
- 2.56 GB
Measured Estimated VRAM 4 GB+
-
9.4B
zai-org
GLM-4.6V-Flash
- Q4_K_M size
- 6.17 GB
- Memory at 8K
- 7.35 GB
Measured Estimated VRAM 8 GB+
-
753.9B
zai-org
GLM-5.2
- UD-Q4_K_M size
- 466 GB
- Memory at 8K
- 467 GB
Measured Estimated
Models of a similar size (other developers)
-
27.3B
Qwen
Qwen3.8-27B
- UD-Q4_K_M size
- 16.5 GB
- Memory at 8K
- 18.2 GB
Measured Estimated VRAM 24 GB+
-
35.5B
huihui-ai
Huihui-Qwen3.6-35B-A3B-abliterated-MTP
- Q2_K size
- 13.2 GB
- Memory at 8K
- 14.5 GB
Measured Estimated VRAM 16 GB+
-
35.5B
ornith-ai
Ornith-1.5-35B-A3B
- Q4_K_M size
- 21.7 GB
- Memory at 8K
- 23.0 GB
Measured Estimated VRAM 24 GB+
-
27.3B
huihui-ai
Huihui-Qwen3.8-27B-abliterated
- Q4_K_M size
- 16.6 GB
- Memory at 8K
- 18.3 GB
Measured Estimated VRAM 24 GB+
-
25.2B
farbodtavakkoli
OTel-2.0-Decision-26B-A4B-4bit
- Q4_K_M size
- 16.8 GB
- Memory at 8K
- 18.4 GB
Measured Estimated VRAM 24 GB+
-
27.3B
JonathanColetti
Qwen3.8-27B-Uncensored
- Q4_K_M size
- 16.5 GB
- Memory at 8K
- 18.3 GB
Measured Estimated VRAM 24 GB+
Frequently asked questions
- How much memory do I need to run GLM-4.7-Flash (Q4_K_M)?
- The file size is 18.3 GB (measured). With a 8,192-token context, the estimated memory including the KV cache is 19.6 GB. The formula is published at https://localllm.az-arte.com/en/guide/memory-calculation.
- Can GLM-4.7-Flash be used commercially?
- The original model's license is MIT License (Commercial use allowed).Allows commercial use, modification and redistribution. When redistributing, keep the copyright notice and license text. Check the original license text before deciding.
- Does GLM-4.7-Flash run in 8 GB, 12 GB, 16 GB or 24 GB of VRAM?
- 8 GB: no quantization fits; 12 GB: up to Q2_K_L (est. 12.7 GB); 16 GB: up to Q3_K_M (est. 15.9 GB); 24 GB: up to UD-Q5_K_XL (est. 23.0 GB) (estimated at 8,192 tokens with an f16 KV cache, files from the most-downloaded source).
- What is the command to run GLM-4.7-Flash with Ollama?
- Run "ollama run hf.co/unsloth/GLM-4.7-Flash-GGUF:Q4_K_M" to use the GGUF on Hugging Face directly. To use another quantization, replace the name at the end.