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

Libraries

How to use ornith-ai/Ornith-1.0-35B 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")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("ornith-ai/Ornith-1.0-35B")
model = AutoModelForMultimodalLM.from_pretrained("ornith-ai/Ornith-1.0-35B", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ornith-ai/Ornith-1.0-35B 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"
# 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",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

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

SGLang

How to use ornith-ai/Ornith-1.0-35B 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" \
    --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",
		"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" \
        --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",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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

Qwen35b-Agent-R2O3: SVD-extracted Ornith LoRA fusion results

#28

by hotdogs - opened Jul 14

Discussion

hotdogs

Jul 14

•

edited Jul 14

🦅 Thank you for Ornith-1.0-35B!

First of all, thank you to the DeepReinforce team for releasing Ornith-1.0-35B under MIT license. It is an incredible model with strong algorithm and reasoning capabilities.

🧬 What We Did

We used SVD Weight-Diff extraction to distill Ornith's unique knowledge into a LoRA adapter (r=32) and merge it with our existing Agent-R2 model (which itself is a 7-LoRA fusion on Qwen-AgentWorld-35B-A3B).

The process:

Computed weight diff between Ornith-1.0-35B and the shared base (Huihui-Qwen-AgentWorld-35B-A3B-abliterated)
Applied SVD decomposition to extract compact LoRA weights
Merged into Agent-R2 at scale α=0.3 (preserving 70% of R2 while adding 30% Ornith influence)

Result: Qwen35b-Agent-R2O3 — a model that retains R2's tool-calling + agent capabilities while benefiting from Ornith's algorithm strength.

📊 Observed Improvements

Hard algorithm tasks: noticeably better on tasks like Raft consensus, vector DB HNSW, and complex load balancing (compared to R2 alone)
Tool calling: fully preserved (R2's 7 LoRAs intact)
No regression: tested on 10+ agentic tasks with no quality loss in conversation, routing, or format compliance

🤝 Sharing Back

We are happy to share our resulting model with the community:

Qwen35b-Agent-R2O3 → https://huggingface.co/hotdogs/Qwen35b-agent-R2O3

GGUF Q4_K_M and Q6_K available for llama.cpp users.

Thanks again for your amazing work! ❤️

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