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

Use Docker

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

LM Studio
Jan

vLLM

How to use bloomer010/Ling-3.0-flash-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-flash-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-flash-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

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

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

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

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

Run Pi

# Start Pi in your project directory:
pi

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

Lemonade

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

Pull the model

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

Run and chat with the model

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

List all available models

lemonade list

Hermes Agent

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

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use bloomer010/Ling-3.0-flash-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-flash-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-flash-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-flash-GGUF / ling3-run.sh

bloomer010

Add ling3-run.sh launcher (auto quant pick, recommended sampling, optional MTP)

92e61a0 verified 30 days ago

Raw

History Blame

3.53 kB

	#!/usr/bin/env bash
	# ling3-run.sh — one-command server launcher for bloomer010 Ling-3.0 GGUFs
	#
	# Picks a quant to fit available memory, applies the source model's recommended
	# sampling, and optionally enables the bundled MTP drafter (flash only).
	#
	# Requires a llama.cpp build with bailingmoe3 support (upstream PR #26608 or
	# newer; older builds: https://github.com/aetherbird/llama.cpp/tree/bailingmoe3-support).
	#
	# Usage:
	# ./ling3-run.sh flash # auto quant, auto memory detection
	# ./ling3-run.sh flash Q4_K_S # explicit quant
	# ./ling3-run.sh flash --mtp # force MTP drafter on
	# ./ling3-run.sh tiny
	# ./ling3-run.sh flash --ctx 65536 --port 8081
	#
	# Env overrides: LLAMA_BIN_DIR (path to llama-server), VRAM_GB / RAM_GB (skip detection)

	set -euo pipefail

	MODEL="${1:-flash}"; shift \|\| true
	QUANT="auto"; MTP="auto"; CTX="32768"; PORT="8080"
	while [ $# -gt 0 ]; do
	case "$1" in
	--mtp) MTP="on"; shift ;;
	--no-mtp) MTP="off"; shift ;;
	--ctx) CTX="$2"; shift 2 ;;
	--port) PORT="$2"; shift 2 ;;
	*) QUANT="$1"; shift ;;
	esac
	done

	REPO="bloomer010/Ling-3.0-${MODEL}-GGUF"
	case "$MODEL" in
	flash)
	# ladder: (min usable total memory GB -> quant), weights + context headroom
	LADDER=(
	"192 UD-Q8_K_XL"
	"136 Q8_0"
	"124 UD-Q6_K_XL"
	"80 Q5_K_M"
	"64 Q4_K_M"
	"56 MXFP4_MOE" # Blackwell/GB10 native; others dequant fallback
	"56 Q4_K_S"
	"48 Q3_K_M"
	"32 UD-Q2_K_XL"
	"24 IQ1_M"
	)
	TEMP="0.6" # source model card recommendation
	;;
	tiny)
	LADDER=(
	"16 BF16"
	"12 UD-Q8_K_XL"
	"10 Q8_0"
	"8 UD-Q6_K_XL"
	"6 Q4_K_M"
	"5 MXFP4_MOE"
	"4 Q3_K_M"
	"3 IQ2_M"
	)
	TEMP="1.0" # source model card recommendation
	;;
	*) echo "unknown model '$MODEL' (flash\|tiny)" >&2; exit 1 ;;
	esac

	# --- memory detection -------------------------------------------------------
	detect_mem() {
	if [ -n "${VRAM_GB:-}" ] && [ -n "${RAM_GB:-}" ]; then
	echo $(( VRAM_GB > RAM_GB ? VRAM_GB : RAM_GB ))
	return
	fi
	local vram ram
	vram=$(nvidia-smi --query-gpu=memory.total --format=csv,noheader,nounits 2>/dev/null \
	\| awk '{s+=$1} END {printf "%d", s/1024}' \|\| echo 0)
	ram=$(free -g \| awk '/^Mem:/ {printf "%d", $2}')
	echo $(( vram > ram ? vram : ram ))
	}

	TOTAL_MEM=$(detect_mem)
	# reserve headroom for KV cache, runtime, and OS
	TOTAL_MEM=$((TOTAL_MEM > 10 ? TOTAL_MEM - 8 : TOTAL_MEM))

	if [ "$QUANT" = "auto" ]; then
	for rung in "${LADDER[@]}"; do
	read -r need q <<< "$rung"
	if [ "$TOTAL_MEM" -ge "$need" ]; then QUANT="$q"; break; fi
	done
	[ "$QUANT" = "auto" ] && QUANT="$(read -r _ q <<< "${LADDER[-1]}"; echo "$q")"
	echo "[ling3] ${TOTAL_MEM} GB detected -> ${QUANT} (override: ./ling3-run.sh ${MODEL} <quant>)"
	fi

	# --- binary -----------------------------------------------------------------
	BIN="${LLAMA_BIN_DIR:-}/llama-server"
	command -v "$BIN" >/dev/null 2>&1 \|\| BIN="llama-server"
	command -v "$BIN" >/dev/null 2>&1 \|\| { echo "llama-server not found (set LLAMA_BIN_DIR)" >&2; exit 1; }

	# --- MTP: on for flash unless disabled; never for tiny (no bundled MTP) ----
	SPEC_ARGS=()
	if [ "$MODEL" = "flash" ] && { [ "$MTP" = "on" ] \|\| [ "$MTP" = "auto" ]; }; then
	SPEC_ARGS=(--spec-type draft-mtp)
	fi

	exec "$BIN" \
	-hf "${REPO}:${QUANT}" \
	--host 0.0.0.0 --port "$PORT" \
	-c "$CTX" -ngl auto -fa on --jinja \
	--temp "$TEMP" --top-p 0.95 --top-k 20 \
	"${SPEC_ARGS[@]}" "$@"