Instructions to use singulared/Ornith-1.0-35B-MTP-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 singulared/Ornith-1.0-35B-MTP-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 singulared/Ornith-1.0-35B-MTP-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf singulared/Ornith-1.0-35B-MTP-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 singulared/Ornith-1.0-35B-MTP-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf singulared/Ornith-1.0-35B-MTP-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 singulared/Ornith-1.0-35B-MTP-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf singulared/Ornith-1.0-35B-MTP-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 singulared/Ornith-1.0-35B-MTP-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf singulared/Ornith-1.0-35B-MTP-GGUF:Q4_K_M

Use Docker

docker model run hf.co/singulared/Ornith-1.0-35B-MTP-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use singulared/Ornith-1.0-35B-MTP-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "singulared/Ornith-1.0-35B-MTP-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": "singulared/Ornith-1.0-35B-MTP-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/singulared/Ornith-1.0-35B-MTP-GGUF:Q4_K_M

Ollama
How to use singulared/Ornith-1.0-35B-MTP-GGUF with Ollama:
```
ollama run hf.co/singulared/Ornith-1.0-35B-MTP-GGUF:Q4_K_M
```

Unsloth Studio

How to use singulared/Ornith-1.0-35B-MTP-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 singulared/Ornith-1.0-35B-MTP-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 singulared/Ornith-1.0-35B-MTP-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for singulared/Ornith-1.0-35B-MTP-GGUF to start chatting

How to use singulared/Ornith-1.0-35B-MTP-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf singulared/Ornith-1.0-35B-MTP-GGUF:Q4_K_M

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": "singulared/Ornith-1.0-35B-MTP-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use singulared/Ornith-1.0-35B-MTP-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 singulared/Ornith-1.0-35B-MTP-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 singulared/Ornith-1.0-35B-MTP-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use singulared/Ornith-1.0-35B-MTP-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf singulared/Ornith-1.0-35B-MTP-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 "singulared/Ornith-1.0-35B-MTP-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"

Docker Model Runner
How to use singulared/Ornith-1.0-35B-MTP-GGUF with Docker Model Runner:
```
docker model run hf.co/singulared/Ornith-1.0-35B-MTP-GGUF:Q4_K_M
```

Lemonade

How to use singulared/Ornith-1.0-35B-MTP-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull singulared/Ornith-1.0-35B-MTP-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Ornith-1.0-35B-MTP-GGUF-Q4_K_M

List all available models

lemonade list

Unquantized weights?

by LeePapa - opened 15 days ago

Discussion

LeePapa

15 days ago

I would like to quantize this into mlx format for apple silicon. I see you have only made available quantized ggufs. I can download these, unquantize them, and then requantized them but i think this will introduce additional quality degradation

singulared

Owner 13 days ago

Hugging Face CLI

MLX does support MTP speculative decoding (mlx-lm PR #990 and the MTPLX project), so keeping the head is worthwhile — but converting these GGUFs isn't the right path, for two reasons:

GGUF→MLX is lossy by construction. MLX expects HF safetensors, so a GGUF→MLX conversion dequantizes to FP16 and then requantizes — exactly the double-quant degradation you want to avoid.
MLX reads the MTP head from the safetensors config (num_nextn_predict_layers), not from a GGUF nextn block — so the grafted head in these GGUFs wouldn't carry into MLX's MTP path anyway.

The clean path is at the safetensors level:

Base Ornith (unquantized, MIT): deepreinforce-ai/Ornith-1.0-35B — note it has no MTP head.
The head I grafted comes from Qwen/Qwen3.6-35B-A3B (Apache-2.0), which ships a native MTP head (Ornith is a qwen35moe fine-tune of it, which is why the graft transfers cleanly).
So: graft Qwen3.6-35B-A3B's MTP head onto Ornith at the safetensors level, then run mlx_lm.convert with your target quant — one clean pass, MTP preserved.

I've only produced GGUF grafts here (for llama.cpp), so I don't have a safetensors+MTP build to point you at, but the two public bases above let you do it directly. Good luck!

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