No official torrent yet — requesting it tells the flock what to host next. You cannot seed a catalog-only page.
Cataloged from DhruvalLabs/GLM-4.7-Flash-GGUF
All models by DhruvalLabs →Quantized GGUF versions of zai-org/GLM-4.7-Flash
Works with llama.cpp · Ollama · LM Studio · Open WebUI · Jan
Can you run a powerful model on a laptop without losing its intelligence?
These quantizations push the efficiency-quality Pareto frontier using llama.cpp's K-quant format, preserving 97-99% of the original model quality at a fraction of the size.
GLM-4.7-Flash-Q4_K_M.gguf — best balance of size and qualityGLM-4.7-Flash-Q8_0.gguf — near-original qualityGLM-4.7-Flash-Q3_K_M.gguf — fits in 8GB with room to spareRun python benchmark.py --model GLM-4.7-Flash to generate speed results.
Run kaggle_bench.ipynb on Kaggle to benchmark this model.
Search for Dhptl/GLM-4.7-Flash in the model browser.
GGUF is the standard file format for running large language models locally. Quantization reduces the number of bits per weight:
Found an issue? Have a question? Open a Discussion in the Community tab above.
If these quantizations were useful, please consider:
ollama run dhptl/glm-4.7-flash# Download the binary from https://github.com/ggerganov/llama.cpp/releases./llama-cli \ -m GLM-4.7-Flash-Q4_K_M.gguf \ -p "You are a helpful assistant." \ --conversation \ -n 512from llama_cpp import Llama
llm = Llama( model_path="./GLM-4.7-Flash-Q4_K_M.gguf", n_gpu_layers=-1, # -1 = offload everything to GPU n_ctx=4096,)
response = llm.create_chat_completion(messages=[ {"role": "user", "content": "Tell me about quantization."}])print(response["choices"][0]["message"]["content"])