Instructions to use WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use WaylonJBrown/ling-3.0-tiny-sft-agentic-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 WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF:Q4_K_M

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 WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF:Q4_K_M

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 WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF:Q4_K_M

Use Docker

docker model run hf.co/WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF:Q4_K_M

Llama new
LM Studio
Jan

vLLM

How to use WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF:Q4_K_M

Ollama
How to use WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF with Ollama:
```
ollama run hf.co/WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF:Q4_K_M
```
Unsloth Desktop

How to use WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF: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": "WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF with Docker Model Runner:
```
docker model run hf.co/WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF:Q4_K_M
```

Lemonade

How to use WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.ling-3.0-tiny-sft-agentic-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use WaylonJBrown/ling-3.0-tiny-sft-agentic-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 WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF: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 WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF: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 "WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF: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"

Ling-3.0-tiny SFT GGUF (agentic SWE)

Merged LoRA + GGUF quants of WaylonJBrown/ling-3.0-tiny-sft-agentic on inclusionAI/Ling-3.0-tiny (7.9B hybrid MoE, ~1.3B active, architecture bailingmoe3).

Converted with k2horizon/llama.cpp (convert_hf_to_gguf.py --outtype bf16, then llama-quantize). Upstream llama.cpp without bailingmoe3 will not load these files.

Files

File	Size	Notes
`Ling-3.0-tiny-sft-agentic-Q4_K_M.gguf`	4.5 GB	Default local quant. MLA `attn_k_b` tensors (ncols 128) fall back to Q5_0.
`Ling-3.0-tiny-sft-agentic-Q8_0.gguf`	7.9 GB	Higher fidelity.
`Ling-3.0-tiny-sft-agentic-BF16.gguf`	15 GB	Source of the quants; not uploaded here.

Run

llama-cli -m Ling-3.0-tiny-sft-agentic-Q4_K_M.gguf \
  --single-turn --no-jinja \
  -p 'Write a Python function that returns Fibonacci numbers.' \
  -n 128 --temp 0.2

Chat template from the base model is embedded in the GGUF. For chat-style turns, drop --no-jinja and use the model's template. SFT was trained with enable_thinking=False.

Training mix (from the adapter)

Source	Count in mix
WlnJBrn09/SFT-1_10-1-26 (SFT-T, upsample 4×)	2,236
nvidia/Nemotron-SFT-SWE-v3.5	5,115
nvidia/Nemotron-RL-Agentic-SWE-Pivot-v1 (`pass_rate >= 0.375`, cap 8k)	8,000
Train total	15,351

LoRA r=16, alpha=32, 1000 steps on 1× H100 (~0.52 epoch), seq 4096, assistant-only labels.

Model tree for WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF

Base model

inclusionAI/Ling-3.0-tiny

Quantized

(38)

this model

WaylonJBrown
/

ling-3.0-tiny-sft-agentic-GGUF

Ling-3.0-tiny SFT GGUF (agentic SWE)

Files

Run

Training mix (from the adapter)

Links

Model tree for WaylonJBrown/ling-3.0-tiny-sft-agentic-GGUF