Instructions to use ccore/Qwen3.5-4B-3D-Coder with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use ccore/Qwen3.5-4B-3D-Coder 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 ccore/Qwen3.5-4B-3D-Coder:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf ccore/Qwen3.5-4B-3D-Coder:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf ccore/Qwen3.5-4B-3D-Coder:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf ccore/Qwen3.5-4B-3D-Coder:Q4_K_M

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf ccore/Qwen3.5-4B-3D-Coder:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama cli -hf ccore/Qwen3.5-4B-3D-Coder:Q4_K_M

Use Docker

docker model run hf.co/ccore/Qwen3.5-4B-3D-Coder:Q4_K_M

Llama new
LM Studio
Jan
Ollama
How to use ccore/Qwen3.5-4B-3D-Coder with Ollama:
```
ollama run hf.co/ccore/Qwen3.5-4B-3D-Coder:Q4_K_M
```
Unsloth Desktop

How to use ccore/Qwen3.5-4B-3D-Coder with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ccore/Qwen3.5-4B-3D-Coder:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "ccore/Qwen3.5-4B-3D-Coder:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use ccore/Qwen3.5-4B-3D-Coder with Docker Model Runner:
```
docker model run hf.co/ccore/Qwen3.5-4B-3D-Coder:Q4_K_M
```

Lemonade

How to use ccore/Qwen3.5-4B-3D-Coder with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull ccore/Qwen3.5-4B-3D-Coder:Q4_K_M

Run and chat with the model

lemonade run user.Qwen3.5-4B-3D-Coder-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use ccore/Qwen3.5-4B-3D-Coder with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ccore/Qwen3.5-4B-3D-Coder:Q4_K_M

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 ccore/Qwen3.5-4B-3D-Coder:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use ccore/Qwen3.5-4B-3D-Coder with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ccore/Qwen3.5-4B-3D-Coder:Q4_K_M

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 "ccore/Qwen3.5-4B-3D-Coder:Q4_K_M" \
  --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"

Fit table: Q4_K_M vs Q8_0 at the recommended ~72K context

by localllmrig - opened 9 days ago

Discussion

localllmrig

9 days ago

Fit table for the two GGUFs, since the setup recommended in #6 (~72K context, f16 KV cache, full GPU offload) is what decides whether a given card works.

The KV cache is small here: only 8 of the 32 layers in config.json are full attention (4 KV heads, head_dim 256), so f16 KV is about 32 KB per token: 0.27 GB at 8K, 2.15 GB at 64K and 2.42 GB at 72K. The linear-attention state is fixed-size and tiny. Need ≈ file + KV cache + ~0.8 GB runtime (our estimate).

File	Size	Need @8K	Need @64K	Need @72K	8 GB card	12 GB card
Q4_K_M	2.71 GB	~3.8 GB	~5.7 GB	~5.9 GB	✓ up to ~118K ctx	✓ up to ~240K ctx
Q8_0	4.48 GB	~5.5 GB	~7.4 GB	~7.7 GB	⚠ ~64K max	✓ up to ~186K ctx

(8 GB counted as ~7.4 GB usable, 12 GB as ~11.4 GB, f16 KV.)

So on an 8 GB card, Q4_K_M runs the recommended 72K context with room to spare, while Q8_0 tops out right around 64K with f16 KV. A q8_0 KV cache (-ctk q8_0 -ctv q8_0) gets Q8_0 to 72K on 8 GB too (~6.5 GB), though the author tested with f16. 16 GB cards run either file at the full 262K. Corrections welcome if anyone has measured numbers.

More per-GPU numbers: https://localllmrig.com/vram/8gb?utm_source=huggingface&utm_medium=community&utm_campaign=qwen3-5-4b-3d-coder

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment