Instructions to use giannisan/GLM-5.2-ds4-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use giannisan/GLM-5.2-ds4-gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="giannisan/GLM-5.2-ds4-gguf",
	filename="GLM-5.2-UD-IQ2_XXS_RoutedIQ2XXS_blk78Q2K.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use giannisan/GLM-5.2-ds4-gguf with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf giannisan/GLM-5.2-ds4-gguf:UD-IQ2_XXS_ROUTEDIQ
# Run inference directly in the terminal:
llama cli -hf giannisan/GLM-5.2-ds4-gguf:UD-IQ2_XXS_ROUTEDIQ

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf giannisan/GLM-5.2-ds4-gguf:UD-IQ2_XXS_ROUTEDIQ
# Run inference directly in the terminal:
llama cli -hf giannisan/GLM-5.2-ds4-gguf:UD-IQ2_XXS_ROUTEDIQ

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf giannisan/GLM-5.2-ds4-gguf:UD-IQ2_XXS_ROUTEDIQ
# Run inference directly in the terminal:
./llama-cli -hf giannisan/GLM-5.2-ds4-gguf:UD-IQ2_XXS_ROUTEDIQ

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf giannisan/GLM-5.2-ds4-gguf:UD-IQ2_XXS_ROUTEDIQ
# Run inference directly in the terminal:
./build/bin/llama-cli -hf giannisan/GLM-5.2-ds4-gguf:UD-IQ2_XXS_ROUTEDIQ

Use Docker

docker model run hf.co/giannisan/GLM-5.2-ds4-gguf:UD-IQ2_XXS_ROUTEDIQ

LM Studio
Jan

Ollama

How to use giannisan/GLM-5.2-ds4-gguf with Ollama:

ollama run hf.co/giannisan/GLM-5.2-ds4-gguf:UD-IQ2_XXS_ROUTEDIQ

Unsloth Studio

How to use giannisan/GLM-5.2-ds4-gguf with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for giannisan/GLM-5.2-ds4-gguf to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for giannisan/GLM-5.2-ds4-gguf to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for giannisan/GLM-5.2-ds4-gguf to start chatting

How to use giannisan/GLM-5.2-ds4-gguf with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf giannisan/GLM-5.2-ds4-gguf:UD-IQ2_XXS_ROUTEDIQ

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "giannisan/GLM-5.2-ds4-gguf:UD-IQ2_XXS_ROUTEDIQ"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use giannisan/GLM-5.2-ds4-gguf with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf giannisan/GLM-5.2-ds4-gguf:UD-IQ2_XXS_ROUTEDIQ

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default giannisan/GLM-5.2-ds4-gguf:UD-IQ2_XXS_ROUTEDIQ

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use giannisan/GLM-5.2-ds4-gguf with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf giannisan/GLM-5.2-ds4-gguf:UD-IQ2_XXS_ROUTEDIQ

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "giannisan/GLM-5.2-ds4-gguf:UD-IQ2_XXS_ROUTEDIQ" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Docker Model Runner
How to use giannisan/GLM-5.2-ds4-gguf with Docker Model Runner:
```
docker model run hf.co/giannisan/GLM-5.2-ds4-gguf:UD-IQ2_XXS_ROUTEDIQ
```

Lemonade

How to use giannisan/GLM-5.2-ds4-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull giannisan/GLM-5.2-ds4-gguf:UD-IQ2_XXS_ROUTEDIQ

Run and chat with the model

lemonade run user.GLM-5.2-ds4-gguf-UD-IQ2_XXS_ROUTEDIQ

List all available models

lemonade list

GLM-5.2 GGUF for ds4 (SSD streaming, CUDA)

This is a mirror of the official ds4 GGUF of GLM-5.2 (743B MoE) built by antirez and published at antirez/GLM-5.2-GGUF (bit-identical file, same sha256). Credit for the quantization is his; this repo re-documents it with the full per-tensor recipe below and pairs it with the CUDA/SSD-streaming usage notes. It is the file used by the ds4 inference engine, specifically the glm-local branch which adds the CUDA port, SSD expert streaming optimizations, and the first MTP speculative-decoding implementation for GLM 5.2 on any backend.

The design target is machines that cannot hold the model in RAM at all: routed experts are read from disk per token while the ~20 GiB of attention/shared weights stay resident. On an RTX 4060 Ti 16GB + 30GB RAM + a Gen4 NVMe it decodes at ~0.4 tokens/s and climbs with disk bandwidth. Yes, that is slow. It is also a 743B model answering on hardware worth less than a mid-range gaming rig.

Files

File	Size	Routed experts (ffn_{gate,up,down}_exps)	Everything else
GLM-5.2-UD-IQ2_XXS_RoutedIQ2XXS_blk78Q2K.gguf	196.6 GiB	IQ2_XXS (layers 3-77, 225 tensors, 177 GiB); blk.78 MTP layer at Q2_K	Q8_0 attention/shared-expert/embeddings/output (872 tensors) + F32 norms (709 tensors), 19.6 GiB

The MTP draft head (blk.78: full GLM layer + nextn eh_proj/enorm/hnorm/shared_head_norm) is included in the main file. No separate draft gguf is needed: pass the same file to --mtp.

Quantization recipe

The filename is the spec. In detail:

Tensor class	Quant	Notes
blk.*.ffn_{gate,up,down}_exps (layers 3-77)	IQ2_XXS	routed experts, uniform on purpose: the streaming expert cache uses fixed-size slabs and the dp4a decode kernels read IQ2_XXS directly
blk.78.ffn_{gate,up,down}_exps	Q2_K	the MTP draft layer's experts; never runs in the main decode loop, only feeds speculation
blk.*.ffn_{gate,up,down}_shexp	Q8_0	shared experts
blk.{0,1,2}.ffn_{gate,up,down}	Q8_0	leading dense layers
blk.*.attn_q_a, attn_q_b, attn_kv_a_mqa, attn_k_b, attn_v_b, attn_output	Q8_0	all MLA attention projections
blk.*.indexer.attn_q_b, indexer.attn_k	Q8_0	DSA sparse-indexer projections
blk.*.indexer.proj, indexer.k_norm(+bias)	F32	indexer scoring head
blk..ffn_gate_inp (router), blk..exp_probs_b (router bias)	F32	learned router, kept exact
blk.78.nextn.eh_proj	Q8_0	MTP embed/hidden fusion
blk.78.nextn.{enorm,hnorm,shared_head_norm}	F32	MTP glue norms
token_embd.weight, output.weight	Q8_0	embeddings and output head
all *_norm.weight	F32

The motivation behind the asymmetry: the routed experts are the majority of the parameter count but each individual expert handles only a fraction of tokens, so aggressive quantization on them costs less in average quality than the same treatment of the router, projections, or shared experts. Keeping the decision-making components at Q8_0 preserves model behavior; crushing the experts buys the size.

SSD streaming adds a second reason: the Q8_0/F32 set (~20 GiB) is resident, so its bytes are paid once in RAM, while the experts are read from disk again and again. Quantizing the experts harder is a per-token bandwidth win; quantizing the resident set harder would save only idle memory. The asymmetry follows the traffic, not just the parameter count.

Usage

Needs the glm-local branch of ds4 (CUDA, sm_89 tested) and a fast NVMe. Expert reads are O_DIRECT through io_uring; host RAM is used for a popularity (LFU) expert cache, so give it whatever you can spare with DS4_CUDA_HOST_EXPERT_CACHE_GB.

git clone -b glm-local https://github.com/giannisanni/neutronstar
cd ds4 && make cuda CUDA_ARCH=sm_89
DS4_GLM_CUDA_UNSAFE=1 DS4_CUDA_HOST_EXPERT_CACHE_GB=7 DS4_CUDA_PARALLEL_FETCH_THREADS=16 \
./ds4 -m GLM-5.2-UD-IQ2_XXS_RoutedIQ2XXS_blk78Q2K.gguf \
  --cuda --ssd-streaming --ssd-streaming-cache-experts 64 \
  --ctx 4096 --tokens 400 --nothink -p "Tell me something surprising about Suriname."

Interactive chat: drop -p. MTP probe telemetry: add --mtp with DS4_MTP_PROBE=1 and DS4_MTP_STREAMING_UNSAFE=1.

Measured on RTX 4060 Ti 16GB / 30GB DDR5 / Gen4 x4 NVMe: prefill ~~0.35 t/s, generation ~0.40 t/s with a 7 GiB host expert cache (~~30% hit rate: the hottest 4% of experts serve 30% of lookups). The engine runs at ~89% of the PCIe link ceiling; a faster disk moves the number almost linearly.

sha256: a49de64c5020432bdae23de36a423a9660a5621bc0db8d12b66bd8814b07fea0

License

Inherits the upstream GLM-5.2 model license (zai-org). The quantization recipe and this card: MIT.

Downloads last month: 174

GGUF

Model size

754B params

Architecture

glm-dsa

Hardware compatibility

2-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for giannisan/GLM-5.2-ds4-gguf

Base model

zai-org/GLM-5.2

Quantized

(98)

this model