Instructions to use ornith-ai/Ornith-1.0-35B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ornith-ai/Ornith-1.0-35B-GGUF with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="ornith-ai/Ornith-1.0-35B-GGUF")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("ornith-ai/Ornith-1.0-35B-GGUF", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

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

Use Docker

docker model run hf.co/ornith-ai/Ornith-1.0-35B-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use ornith-ai/Ornith-1.0-35B-GGUF with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/ornith-ai/Ornith-1.0-35B-GGUF:Q4_K_M

SGLang

How to use ornith-ai/Ornith-1.0-35B-GGUF with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "ornith-ai/Ornith-1.0-35B-GGUF" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ornith-ai/Ornith-1.0-35B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "ornith-ai/Ornith-1.0-35B-GGUF" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ornith-ai/Ornith-1.0-35B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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

Unsloth Studio

How to use ornith-ai/Ornith-1.0-35B-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 ornith-ai/Ornith-1.0-35B-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 ornith-ai/Ornith-1.0-35B-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

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

How to use ornith-ai/Ornith-1.0-35B-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ornith-ai/Ornith-1.0-35B-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": "ornith-ai/Ornith-1.0-35B-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

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

Lemonade

How to use ornith-ai/Ornith-1.0-35B-GGUF with Lemonade:

Pull the model

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

Run and chat with the model

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

List all available models

lemonade list

Hermes Agent

How to use ornith-ai/Ornith-1.0-35B-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 ornith-ai/Ornith-1.0-35B-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 ornith-ai/Ornith-1.0-35B-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use ornith-ai/Ornith-1.0-35B-GGUF with OpenClaw:

Start the llama.cpp server

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

Hermes Agent + chat template: tool calls fail silently (XML vs JSON format)

#46

by k-mktr - opened Jul 21

Discussion

k-mktr

Jul 21

•

edited Jul 21

If you are running Ornith with a qwen3.6-froggeric chat template through Hermes Agent and wondering why tool calling does not work - the model outputs XML but Hermes expects JSON.

Root cause

The chat template (qwen3.6-froggeric-v21.3) defaults to XML tool call format:

{%- set _tool_format = tool_call_format if tool_call_format is defined else "xml" %}

The model follows instructions and outputs XML like:

<function=web_search>
<parameter=query>current Bitcoin price</parameter>

But Hermes Agent expects standard OpenAI JSON tool calls. Result: finish_reason: stop instead of tool_calls. The agent never receives tool results and falls back to text-only responses.

Fix

In your Hermes config (~/.hermes/config.yaml), under the model: section, add:

model:
  chat_template_kwargs: '{"tool_call_format": "json"}'
  extra_body: '{"chat_template_kwargs": {"tool_call_format": "json"}}'

This tells the template engine to use JSON format instead of XML. The template supports this natively.

Alternative fix (server-side)

Change the default in chat_template.jinja itself from xml to json on line 2. This makes it work for all clients without client-side config.

Why this matters

This is a silent failure - no errors, no warnings. The model appears to work but never actually uses tools. If you see tool_turns=0 in Hermes agent logs despite tools being available, this is likely why.

The same mechanism works for other template parameters: enable_thinking, auto_disable_thinking_with_tools, preserve_thinking, max_tool_arg_chars, max_tool_response_chars.

k-mktr

Jul 21

This comment has been hidden (marked as Spam)

k-mktr

Jul 21

This comment has been hidden

veldierin

19 days ago

•

edited 19 days ago

Is this really the case though?

I have been using the hermes agent since May and been using the older v20 XML only froggeric chat template with both qwen3.6-27b/qwen3.6-35b-a3b via GGUF + the latest beta llama.cpp and have not run into problems.

the newest v21.3 froggeric that splits json/xml has been a nightmare though. i made edits and removed all reference to JSON and my hermes agent is just fine doing large 20-30+ toolcall oneshots for briefing crons with the standard v21.3 froggeric chat template using XML now. I've been running the v21.3 with my edits for a week now and only 1x time out of >42 briefing crons, each executing 5-30+ tool calls in a single one-shot cron session over a week (6+ crons per day), did i have a problem.

edit to clarify, i suppose this may only be a llama.cpp behavior which handles XML better. the whole point of that JSON addition to my recollection was to support vLLM.

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