GLM-4.7-Flash ยท GGUF (llama.cpp)
GGUF quantizations of zai-org/GLM-4.7-Flash for llama.cpp / Ollama / LM Studio. Produced by axteam, converted with the official llama.cpp convert_hf_to_gguf.py (the original chat_template is embedded).
Revision 2026-11-02: the
Q3_K_MandQ2_Kfiles were removed from this repo:Q3_K_Mbecause its perplexity check gave a result that could not be explained,Q2_Kbecause it failed the chat check (it never answered the Korean question in Korean, only repeating English commentary until the 256-token limit). The other files are unchanged.
Files
| File | Size | Notes |
|---|---|---|
GLM-4.7-Flash-Q4_K_M.gguf |
18.9 GB | 4-bit k-quant (medium) โ recommended size/quality balance. |
GLM-4.7-Flash-Q5_K_M.gguf |
22.2 GB | 5-bit k-quant (medium) โ high quality. |
GLM-4.7-Flash-Q6_K.gguf |
25.7 GB | 6-bit k-quant โ high quality, larger. |
GLM-4.7-Flash-Q8_0.gguf |
33.2 GB | 8-bit โ closest to the original. |
Quantization method
These GGUF files are post-training quantization (PTQ) made with llama.cpp (convert_hf_to_gguf.py โ llama-quantize): the BF16 weights are quantized block-wise (k-quants: per-block scales with mixed bit allocation), with no retraining. The original chat_template is embedded in each file.
Requirements
| llama.cpp | build โฅ b10251 โ architecture deepseek2 + MTP layer needs โฅ b10251, and the commands on this card use -fa on (โฅ b6325). Run here on 2026-11-02 with b12005 (commit e0f142d, CUDA build) on an NVIDIA B200: every file in this repo loaded with all layers on the GPU and two short chat questions (capital of France / of Korea) were answered correctly with every file. Perplexity was measured too but is not published: a Q3_K_M file, since removed from this repo, scored far below Q8_0 on WikiText-2 (with both 16 and 64 windows) for a reason that was not found, so the perplexity measurement is not used for this model. |
| Memory | the chosen GGUF file (sizes in the table) plus KV cache for the context (-c); --cache-type-k/v q8_0 -fa on halves the KV cache. |
| Input | text. |
| Ollama / LM Studio | not tested here. |
Run with llama.cpp
# pull one quant straight from the Hub and serve an OpenAI-compatible API on :8080
llama-server -hf axteam/GLM-4.7-Flash-GGUF:Q4_K_M -c 8192 --jinja
# or download a single file and chat in the terminal
hf download axteam/GLM-4.7-Flash-GGUF GLM-4.7-Flash-Q4_K_M.gguf --local-dir .
llama-cli -m GLM-4.7-Flash-Q4_K_M.gguf -c 8192 --jinja
Query the server:
curl http://localhost:8080/v1/chat/completions -H "Content-Type: application/json" -d '{
"messages": [{"role": "user", "content": "๋ํ๋ฏผ๊ตญ์ ์๋๋ ์ด๋์ธ๊ฐ์?"}],
"temperature": 1.0, "max_tokens": 512
}'
Reduce KV-cache memory (optional, run-time)
KV-cache type is chosen at run time in llama.cpp (it is not baked into the file):
llama-server -hf axteam/GLM-4.7-Flash-GGUF:Q4_K_M -c 8192 --jinja --cache-type-k q8_0 --cache-type-v q8_0 -fa on
Ollama
ollama run hf.co/axteam/GLM-4.7-Flash-GGUF:Q4_K_M
LM Studio
Search for axteam/GLM-4.7-Flash-GGUF in the model browser and pick a quant (e.g. Q4_K_M).
Recommended sampling (from the original
generation_config.json):temperature=1.0.
License
Quantized derivative of zai-org/GLM-4.7-Flash (MIT). Original weights & tokenizer belong to Z.ai / zai-org.
ํ๊ตญ์ด ์์ฝ
zai-org/GLM-4.7-Flash์ GGUF ์์ํ ๋น๋ (llama.cppยทOllamaยทLM Studio์ฉ) ยท ์ ์ axteam
๊ฐ์
| ํญ๋ชฉ | ๋ด์ฉ |
|---|---|
| ์๋ณธ ๋ชจ๋ธ | zai-org/GLM-4.7-Flash |
| ํ์ | GGUF (llama.cpp) |
| ๋ณํ ๋๊ตฌ | llama.cpp convert_hf_to_gguf.py โ llama-quantize |
| ํ์ผ ์ | 4๊ฐ |
์์ํ ์์ธ
- ์์ํ ์ ์ฐจ: PTQ(ํ์ต ํ ์์ํ, ์ฌํ์ต ์์), llama.cpp ๋ธ๋ก ๋จ์ ์์ํ(k-quant
Q*_K,Q8_0) - chat template: GGUF ํ์ผ์ ๋ด์ฅ
- ํ์ง: PPL์ ์์ผ๋ ์ฃ์ง ์์ โ ์ดํ ์ด ๋ฆฌํฌ์์ ์ญ์ ํ
Q3_K_Mํ์ผ์ด ์์ธ์ ์ฐพ์ง ๋ชปํ ์ฑQ8_0๋ณด๋ค ํฌ๊ฒ ๋ฎ๊ฒ ๋์ ์ด ๋ชจ๋ธ์ PPL ์ธก์ ์ ์ฐ์ง ์์ - ์์ ์ด๋ ฅ (2026-11-02):
Q3_K_MยทQ2_Kํ์ผ ์ญ์ โQ3_K_M์ PPL ์ ๊ฒ ๊ฒฐ๊ณผ๋ฅผ ์ค๋ช ํ ์ ์์๊ณ ,Q2_K๋ ์ง๋ฌธ ์ ๊ฒ์์ ํ๊ตญ์ด ์ง๋ฌธ์ ํ๊ตญ์ด๋ก ๋ตํ์ง ๋ชปํ๊ณ ์์ด ์ค๋ช ๋ฌธ๋ง 256ํ ํฐ ํ๋๊น์ง ๋ํ์ดํจ, ๋๋จธ์ง ํ์ผ์ ๊ทธ๋๋ก
์ ๊ณต ํ์ผ
| ํ์ผ | ํฌ๊ธฐ | ์ค๋ช |
|---|---|---|
GLM-4.7-Flash-Q4_K_M.gguf |
18.9 GB | 4๋นํธ k-quant โ ํฌ๊ธฐยทํ์ง ๊ท ํ (๊ถ์ฅ) |
GLM-4.7-Flash-Q5_K_M.gguf |
22.2 GB | 5๋นํธ k-quant โ ๊ณ ํ์ง |
GLM-4.7-Flash-Q6_K.gguf |
25.7 GB | 6๋นํธ k-quant โ ๊ณ ํ์ง, ์ฉ๋ ํผ |
GLM-4.7-Flash-Q8_0.gguf |
33.2 GB | 8๋นํธ โ ์๋ณธ์ ๊ฐ์ฅ ๊ฐ๊น์ |
์๊ตฌ์ฌํญ
- llama.cpp: b10251 ์ด์ (
deepseek2๊ตฌ์กฐ + MTP ์ธต ์ง์ b10251๋ถํฐ, ์์ ๋ช ๋ น์-fa on์ต์ b6325๋ถํฐ) - ์คํ ํ์ธ: 2026-11-02 b12005(์ปค๋ฐ e0f142d) CUDA ๋น๋๋ก NVIDIA B200์์ ์ด ๋ฆฌํฌ์ ๋ชจ๋ ํ์ผ์ ์ ์ธต GPU์ ์ฌ๋ ค ์คํ โ ๋ชจ๋ ํ์ผ์์ ์งง์ ์ง๋ฌธ 2๊ฐ(ํ๋์คยท๋ํ๋ฏผ๊ตญ ์๋)์ ๋ฐ๋ฅด๊ฒ ๋ตํจ, PPL๋ ์์ผ๋ ์ฃ์ง ์์(์ดํ ์ด ๋ฆฌํฌ์์ ์ญ์ ํ
Q3_K_Mํ์ผ์ด ์์ธ์ ์ฐพ์ง ๋ชปํ ์ฑ ์์ด ํ ์คํธ์์ 16ยท64์ฒญํฌ ๋ชจ๋Q8_0๋ณด๋ค ํฌ๊ฒ ๋ฎ๊ฒ ๋์ ์ด ๋ชจ๋ธ์ PPL ์ธก์ ์ ์ฐ์ง ์์) - ๋ฉ๋ชจ๋ฆฌ: ์ ํํ GGUF ํ์ผ ํฌ๊ธฐ + ์ปจํ
์คํธ(
-c)๋งํผ์ KV ์บ์ (KV ์บ์ ์ค์ด๋ ๋ฐฉ๋ฒ์ ์๋ ์ฌ์ฉ๋ฒ ์ฐธ๊ณ ) - ์ ๋ ฅ: ํ ์คํธ
- OllamaยทLM Studio: ์ ์ ์ธก์์ ์คํ ๋ฏธํ์ธ
์ฌ์ฉ๋ฒ
- ํ์ผ ๋ฐ๊ธฐ:
hf download axteam/GLM-4.7-Flash-GGUF GLM-4.7-Flash-Q4_K_M.gguf --local-dir .(์๋-m๋ช ๋ น์ ํ์, ๋ค๋ฅธ ์์ํ๋ ํ์ผ๋ช ์Q4_K_M๋ง ๋ณ๊ฒฝ) - ์๋ฒ ์คํ:
llama-server -hf axteam/GLM-4.7-Flash-GGUF:Q4_K_M -c 8192 --jinjaโ OpenAI ํธํ API - ํฐ๋ฏธ๋ ๋ํ:
llama-cli -m GLM-4.7-Flash-Q4_K_M.gguf -c 8192 --jinja - Ollama:
ollama run hf.co/axteam/GLM-4.7-Flash-GGUF:Q4_K_M - LM Studio: ๋ชจ๋ธ ๊ฒ์์์
axteam/GLM-4.7-Flash-GGUF์ ๋ ฅ ํ ์์ํ ๋จ๊ณ ์ ํ - KV ์บ์ ์ ๊ฐ(์ ํ):
--cache-type-k q8_0 --cache-type-v q8_0 -fa onโ KV ์บ์ ์ฝ ์ ๋ฐ, GGUF ํ์ผ๊ณผ ์๊ด์์ด ์คํ ์ ์ง์ - ๊ถ์ฅ ์ํ๋ง:
temperature=1.0(์๋ณธgeneration_config.json๊ธฐ์ค)
๋ผ์ด์ ์ค
- ๋ผ์ด์ ์ค: MIT โ ์๋ณธ ๋ชจ๋ธ์ ๋ผ์ด์ ์ค๋ฅผ ๊ทธ๋๋ก ๋ฐ๋ฆ
- ๊ถ๋ฆฌ: ์๋ณธ ๊ฐ์ค์นยทํ ํฌ๋์ด์ ๋ Z.ai ์์ , ์ด ๋ฆฌํฌ๋ ์์ํ ํ์๋ฌผ