Instructions to use sibitim/weft-models with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use sibitim/weft-models 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 sibitim/weft-models:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf sibitim/weft-models:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf sibitim/weft-models:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf sibitim/weft-models: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 sibitim/weft-models:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf sibitim/weft-models: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 sibitim/weft-models:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf sibitim/weft-models:Q4_K_M

Use Docker

docker model run hf.co/sibitim/weft-models:Q4_K_M

LM Studio
Jan
Ollama
How to use sibitim/weft-models with Ollama:
```
ollama run hf.co/sibitim/weft-models:Q4_K_M
```
Unsloth Desktop

How to use sibitim/weft-models with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf sibitim/weft-models: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": "sibitim/weft-models:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use sibitim/weft-models with Docker Model Runner:
```
docker model run hf.co/sibitim/weft-models:Q4_K_M
```

Lemonade

How to use sibitim/weft-models with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull sibitim/weft-models:Q4_K_M

Run and chat with the model

lemonade run user.weft-models-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use sibitim/weft-models with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf sibitim/weft-models: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 sibitim/weft-models:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use sibitim/weft-models with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf sibitim/weft-models: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 "sibitim/weft-models: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"

Weft models

Mirror of the on-device models used by the Weft app, so the app downloads them from one place. Every model keeps its own license, which is in its folder. Only gliner2.5-multi-v1/encoder_int8.onnx is modified; every other file is identical to its source (same SHA-256 as the original repository).

Folder	What	Source	License
`embeddinggemma-2/`	EmbeddingGemma-2, int8 ONNX (search and linking by meaning)	onnx-community/embeddinggemma-2-ONNX, from google/embeddinggemma-2	Apache 2.0
`gliner2.5-multi-v1/`	GLiNER2.5 multi (names and their type): encoder quantized to int8 by Weft, boundary head and tokenizer unchanged	DanKau/gliner2.5-multi-v1-onnx, from fastino/gliner2.5-multi-v1	Apache 2.0
`lfm2.5-1.2b-instruct/`	LFM2.5-1.2B-Instruct, Q4_K_M GGUF (answers)	LiquidAI/LFM2.5-1.2B-Instruct-GGUF	LFM Open License v1.0
`minicpm5-2b/`	MiniCPM5-2B, Q4_K_M GGUF (answers, optional)	openbmb/MiniCPM5-2B-GGUF	Apache 2.0

Changes

gliner2.5-multi-v1/encoder_int8.onnx was changed by Weft. It is the fp32 onnx/encoder.onnx of DanKau/gliner2.5-multi-v1-onnx, quantized to int8 with ONNX Runtime dynamic quantization: 337 MB instead of 1.1 GB, the same entities found, with scores within about 0.08 of the original.

Notices

LFM2.5-1.2B-Instruct is licensed under the LFM Open License v1.0 by Liquid AI. Commercial use is licensed only below the license's revenue threshold; read lfm2.5-1.2b-instruct/LICENSE. "LFM" and "Liquid AI" are names of Liquid AI, used here only to say where the model comes from.
EmbeddingGemma-2 by Google, GLiNER2.5 by Fastino and MiniCPM5 by OpenBMB are under the Apache License 2.0 (the LICENSE file in each folder).

Downloads last month: 6

GGUF

Model size

1B params

Architecture

lfm2

Hardware compatibility

4-bit

Inference Providers NEW

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