Instructions to use bloomer010/Ling-3.0-tiny-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 bloomer010/Ling-3.0-tiny-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 bloomer010/Ling-3.0-tiny-GGUF:UD-Q4_K_XL
# Run inference directly in the terminal:
llama cli -hf bloomer010/Ling-3.0-tiny-GGUF:UD-Q4_K_XL

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf bloomer010/Ling-3.0-tiny-GGUF:UD-Q4_K_XL
# Run inference directly in the terminal:
llama cli -hf bloomer010/Ling-3.0-tiny-GGUF:UD-Q4_K_XL

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 bloomer010/Ling-3.0-tiny-GGUF:UD-Q4_K_XL
# Run inference directly in the terminal:
./llama-cli -hf bloomer010/Ling-3.0-tiny-GGUF:UD-Q4_K_XL

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 bloomer010/Ling-3.0-tiny-GGUF:UD-Q4_K_XL
# Run inference directly in the terminal:
./build/bin/llama-cli -hf bloomer010/Ling-3.0-tiny-GGUF:UD-Q4_K_XL

Use Docker

docker model run hf.co/bloomer010/Ling-3.0-tiny-GGUF:UD-Q4_K_XL

LM Studio
Jan

vLLM

How to use bloomer010/Ling-3.0-tiny-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "bloomer010/Ling-3.0-tiny-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": "bloomer010/Ling-3.0-tiny-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/bloomer010/Ling-3.0-tiny-GGUF:UD-Q4_K_XL

Ollama
How to use bloomer010/Ling-3.0-tiny-GGUF with Ollama:
```
ollama run hf.co/bloomer010/Ling-3.0-tiny-GGUF:UD-Q4_K_XL
```

Unsloth Studio

How to use bloomer010/Ling-3.0-tiny-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 bloomer010/Ling-3.0-tiny-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 bloomer010/Ling-3.0-tiny-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for bloomer010/Ling-3.0-tiny-GGUF to start chatting

How to use bloomer010/Ling-3.0-tiny-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf bloomer010/Ling-3.0-tiny-GGUF:UD-Q4_K_XL

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": "bloomer010/Ling-3.0-tiny-GGUF:UD-Q4_K_XL"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use bloomer010/Ling-3.0-tiny-GGUF with Docker Model Runner:
```
docker model run hf.co/bloomer010/Ling-3.0-tiny-GGUF:UD-Q4_K_XL
```

Lemonade

How to use bloomer010/Ling-3.0-tiny-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull bloomer010/Ling-3.0-tiny-GGUF:UD-Q4_K_XL

Run and chat with the model

lemonade run user.Ling-3.0-tiny-GGUF-UD-Q4_K_XL

List all available models

lemonade list

Hermes Agent

How to use bloomer010/Ling-3.0-tiny-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 bloomer010/Ling-3.0-tiny-GGUF:UD-Q4_K_XL

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 bloomer010/Ling-3.0-tiny-GGUF:UD-Q4_K_XL

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use bloomer010/Ling-3.0-tiny-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf bloomer010/Ling-3.0-tiny-GGUF:UD-Q4_K_XL

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 "bloomer010/Ling-3.0-tiny-GGUF:UD-Q4_K_XL" \
  --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-GGUF / README.md

bloomer010

Update README.md

76d03bf verified 1 day ago

preview code

Raw

History Blame Contribute Delete

4.25 kB

	---
	license: mit
	base_model:
	- inclusionAI/Ling-3.0-tiny
	pipeline_tag: text-generation
	library_name: llama.cpp
	tags:
	- gguf
	- bailingmoe3
	- mixture-of-experts
	- conversational
	---

	# Ling-3.0-tiny GGUF

	GGUF conversions of [inclusionAI/Ling-3.0-tiny](https://huggingface.co/inclusionAI/Ling-3.0-tiny),
	converted directly from the released BF16 safetensors.

	🎉 `bailingmoe3` (including the Q-LoRA attention path) is supported in stock llama.cpp since
	[PR #26608](https://github.com/ggml-org/llama.cpp/pull/26608) (merged 2026-08-17, commit
	`3733366720`). Any build from that commit onward loads these files directly:

	```bash
	llama-server -hf bloomer010/Ling-3.0-tiny-GGUF:Q4_K_M
	```

	## Files

	For tiny models, precision is especially crucial.

	Generally...
	Larger files = more precision.
	More compression = more slop and misbehavin'.

	Use UD-Q8_K_XL for near-full precision performance.

	\| Quant \| Size \| your memory \|
	\| --- \| ---: \| --- \|
	\| BF16 \| 15.8 GB \| 16 GB+ \|
	\| UD-Q8_K_XL \| 11.19 GB \| 12 GB+ \|
	\| Q8_0 \| 8.41 GB \| 10 GB+ \|
	\| UD-Q6_K_XL \| 7.27 GB \| 8 GB+ \|
	\| Q6_K \| 6.50 GB \| 8 GB+ \|
	\| Q5_K_M \| 5.64 GB \| 7 GB+ \|
	\| Q5_K_S \| 5.48 GB \| 6 GB+ \|
	\| Q5_0 \| 5.48 GB \| 6 GB+ \|
	\| Q4_K_M \| 4.82 GB \| 6 GB+ \|
	\| Q4_K_S \| 4.55 GB \| 6 GB+ \|
	\| Q4_0 \| 4.53 GB \| 6 GB+ \|
	\| MXFP4_MOE \| 4.72 GB \| 6 GB+ ¹ \|
	\| IQ4_XS \| 4.29 GB \| 5 GB+ \|
	\| Q3_K_M \| 3.84 GB \| 5 GB+ \|
	\| Q3_K_S \| 3.51 GB \| 5 GB+ \|
	\| IQ3_S \| 3.51 GB \| 4 GB+ \|
	\| IQ3_XXS \| 3.13 GB \| 4 GB+ \|
	\| Q2_K \| 2.99 GB \| 4 GB+ \|
	\| IQ2_M \| 2.70 GB \| 3 GB+ \|
	\| IQ2_S \| 2.48 GB \| 3 GB+ \|
	\| IQ2_XS \| 2.43 GB \| 3 GB+ \|
	\| IQ2_XXS \| 2.21 GB \| 3 GB+ \|
	\| IQ1_M \| 1.93 GB \| 3 GB+ \|
	\| IQ1_S \| 1.76 GB \| 2 GB+ \|
	\| Q1_0 \| 1.30 GB \| 2 GB+ \|

	¹ `MXFP4_MOE` runs its native path on MXFP4-capable GPUs (Blackwell RTX 50-series, GB10/DGX
	Spark). Elsewhere it falls back to a slower dequant path — prefer a K-quant on older hardware.

	## Importance Matrix

	The IQ-quant rungs (`IQ1_S` through `IQ4_XS`) were generated with a model-specific importance
	matrix:

	- Wikitext-2 raw training text
	- 100 chunks
	- 512 tokens per chunk
	- 51,200 calibration tokens total
	- 332 matrix entries

	## XL Quantization Recipes

	`UD-Q8_K_XL` uses Q8_0 for the main expert gate and up tensors. Token embeddings, expert down
	projections, attention and Q-LoRA projections, and KDA projections remain BF16.

	`UD-Q6_K_XL` uses Q6_K for the main expert gate and up tensors. Token embeddings, output weights,
	expert down projections, attention and Q-LoRA projections, and KDA projections use Q8_0. It was
	generated with the importance matrix described above.

	## Architecture

	- 7.9B total parameters and 1.3B active parameters per token
	- 24 layers: 18 KDA layers and 6 MLA layers
	- 128 routed experts, 8 active per token, plus 1 shared expert
	- Q-LoRA rank 256 and KV-LoRA rank 512
	- 131,072-token context in the released configuration
	- No bundled MTP block for this model (`num_nextn_predict_layers: 0`)

	## Validation

	- BF16 conversion completed with 526 tensors, including all 18 Q-LoRA tensors
	- CPU and CUDA architecture tests passed
	- BF16, Q8_0, Q6_K, Q4_K_M, and MXFP4_MOE loaded and generated tokens with CUDA
	- Q1_0, IQ2_M, Q3_K_M, Q5_K_S, and Q5_K_M passed CPU-only prompt processing and token generation
	tests
	- UD-Q6_K_XL and UD-Q8_K_XL passed CPU-only prompt processing and token generation tests
	- IQ1_S, IQ1_M, IQ2_S, IQ2_XS, IQ2_XXS, IQ3_XXS, IQ3_S, IQ4_XS, Q2_K, Q3_K_S, Q4_K_S, Q4_0, and
	Q5_0 passed load and generation tests
	- CUDA testing used an RTX 4070 and RTX 3060

	## Build

	```bash
	git clone https://github.com/ggml-org/llama.cpp.git # bailingmoe3 merged 2026-08-17
	# pre-merge builds:
	# git clone --branch bailingmoe3-support https://github.com/aetherbird/llama.cpp.git
	cd llama.cpp
	cmake -B build -DGGML_CUDA=ON
	cmake --build build --config Release -j --target llama-cli llama-server
	```

	## Usage

	```bash
	./build/bin/llama-server \
	-m Ling-3.0-tiny-Q4_K_M.gguf \
	-c 131072 \
	-ngl auto \
	--flash-attn auto \
	--temp 1.0 --top-p 0.95 --top-k 20 \
	--jinja
	```

	Thinking is enabled by default; disable per request with
	`"chat_template_kwargs": {"enable_thinking": false}`. Recommended sampling parameters from the
	source model card are `temperature=1.0`, `top_p=0.95`, and `top_k=20`.