legraphista's picture
Upload README.md with huggingface_hub
08de942 verified
|
raw
history blame
6 kB
metadata
base_model: THUDM/glm-4-9b-chat
inference: false
language:
  - zh
  - en
library_name: gguf
license: other
license_link: https://huggingface.co/THUDM/glm-4-9b-chat/blob/main/LICENSE
license_name: glm-4
pipeline_tag: text-generation
quantized_by: legraphista
tags:
  - glm
  - chatglm
  - thudm
  - quantized
  - GGUF
  - quantization
  - imat
  - imatrix
  - static
  - 16bit
  - 8bit
  - 6bit
  - 5bit
  - 4bit
  - 3bit
  - 2bit
  - 1bit

glm-4-9b-chat-IMat-GGUF

Llama.cpp imatrix quantization of THUDM/glm-4-9b-chat

Original Model: THUDM/glm-4-9b-chat
Original dtype: BF16 (bfloat16)
Quantized by: https://github.com/ggerganov/llama.cpp/pull/6999#issue-comment-box
IMatrix dataset: here


Files

IMatrix

Status: βœ… Available
Link: here

Common Quants

Filename Quant type File Size Status Uses IMatrix Is Split
glm-4-9b-chat.Q8_0 Q8_0 - ⏳ Processing βšͺ Static -
glm-4-9b-chat.Q6_K Q6_K - ⏳ Processing βšͺ Static -
glm-4-9b-chat.Q4_K Q4_K - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.Q3_K Q3_K - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.Q2_K Q2_K - ⏳ Processing 🟒 IMatrix -

All Quants

Filename Quant type File Size Status Uses IMatrix Is Split
glm-4-9b-chat.BF16 BF16 - ⏳ Processing βšͺ Static -
glm-4-9b-chat.FP16 F16 - ⏳ Processing βšͺ Static -
glm-4-9b-chat.Q8_0 Q8_0 - ⏳ Processing βšͺ Static -
glm-4-9b-chat.Q6_K Q6_K - ⏳ Processing βšͺ Static -
glm-4-9b-chat.Q5_K Q5_K - ⏳ Processing βšͺ Static -
glm-4-9b-chat.Q5_K_S Q5_K_S - ⏳ Processing βšͺ Static -
glm-4-9b-chat.Q4_K Q4_K - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.Q4_K_S Q4_K_S - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.IQ4_NL IQ4_NL - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.IQ4_XS IQ4_XS - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.Q3_K Q3_K - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.Q3_K_L Q3_K_L - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.Q3_K_S Q3_K_S - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.IQ3_M IQ3_M - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.IQ3_S IQ3_S - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.IQ3_XS IQ3_XS - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.IQ3_XXS IQ3_XXS - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.Q2_K Q2_K - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.Q2_K_S Q2_K_S - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.IQ2_M IQ2_M - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.IQ2_S IQ2_S - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.IQ2_XS IQ2_XS - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.IQ2_XXS IQ2_XXS - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.IQ1_M IQ1_M - ⏳ Processing 🟒 IMatrix -
glm-4-9b-chat.IQ1_S IQ1_S - ⏳ Processing 🟒 IMatrix -

Downloading using huggingface-cli

If you do not have hugginface-cli installed:

pip install -U "huggingface_hub[cli]"

Download the specific file you want:

huggingface-cli download legraphista/glm-4-9b-chat-IMat-GGUF --include "glm-4-9b-chat.Q8_0.gguf" --local-dir ./

If the model file is big, it has been split into multiple files. In order to download them all to a local folder, run:

huggingface-cli download legraphista/glm-4-9b-chat-IMat-GGUF --include "glm-4-9b-chat.Q8_0/*" --local-dir ./
# see FAQ for merging GGUF's

Inference

Simple chat template

[gMASK]<sop><|user|>
{user_prompt}<|assistant|>
{assistant_response}<|user|>
{next_user_prompt}

Chat template with system prompt

[gMASK]<sop><|system|>
{system_prompt}<|user|>
{user_prompt}<|assistant|>
{assistant_response}<|user|>
{next_user_prompt}

Llama.cpp

llama.cpp/main -m glm-4-9b-chat.Q8_0.gguf --color -i -p "prompt here (according to the chat template)"

FAQ

Why is the IMatrix not applied everywhere?

According to this investigation, it appears that lower quantizations are the only ones that benefit from the imatrix input (as per hellaswag results).

How do I merge a split GGUF?

  1. Make sure you have gguf-split available
  2. Locate your GGUF chunks folder (ex: glm-4-9b-chat.Q8_0)
  3. Run gguf-split --merge glm-4-9b-chat.Q8_0/glm-4-9b-chat.Q8_0-00001-of-XXXXX.gguf glm-4-9b-chat.Q8_0.gguf
    • Make sure to point gguf-split to the first chunk of the split.

Got a suggestion? Ping me @legraphista!