GLM-4.7-Flash is a highly efficient 30B-A3B Mixture-of-Experts (MoE) model. It is designed to be the strongest model in the 30B parameter class, offering a powerful option for lightweight deployment that perfectly balances performance and efficiency.

Evaluation Results

Benchmark	GLM-4.7-Flash	Qwen3-30B-A3B-Thinking-2507	GPT-OSS-20B
AIME 25	91.6	85.0	91.7
GPQA	75.2	73.4	71.5
LCB v6	64.0	66.0	61.0
HLE	14.4	9.8	10.9
SWE-bench Verified	59.2	22.0	34.0
τ²-Bench	79.5	49.0	47.7
BrowseComp	42.8	2.29	28.3

Files & Quantization

To see the available files, please verify the Files and versions tab.

How to Run (llama.cpp)

Recommended Parameters:

Temperature: 1.0 (Standard) or 0.7 (For stricter adherence)
Top-P: 0.95
Context: -c (Adjust based on available RAM).

CLI Example

bash

./llama-cli -m GLM-4.7-Flash.Q4_K_M.gguf \
  -c 8192 \
  --temp 1.0 \
  --top-p 0.95 \
  -p "User: Write a Python script to calculate Fibonacci numbers.\nAssistant:" \
  -cnv

Server Example

bash

./llama-server -m GLM-4.7-Flash.Q4_K_M.gguf \
  --port 8080 \
  --host 0.0.0.0 \
  -c 16384 \
  -ngl 99

AaryanK/GLM-4.7-Flash-GGUF

GLM-4.7-Flash-GGUF

Description

Evaluation Results

Files & Quantization

How to Run (llama.cpp)

CLI Example

Server Example

AaryanK/
GLM-4.7-Flash-GGUF