Instructions to use ornith-ai/Ornith-1.5-35B-A3B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ornith-ai/Ornith-1.5-35B-A3B with Transformers:

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

pipe = pipeline("text-generation", model="ornith-ai/Ornith-1.5-35B-A3B")
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-1.5-35B-A3B")
model = AutoModelForMultimodalLM.from_pretrained("ornith-ai/Ornith-1.5-35B-A3B", 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-1.5-35B-A3B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ornith-ai/Ornith-1.5-35B-A3B"
# 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-1.5-35B-A3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ornith-ai/Ornith-1.5-35B-A3B

SGLang

How to use ornith-ai/Ornith-1.5-35B-A3B 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-1.5-35B-A3B" \
    --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-1.5-35B-A3B",
		"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-1.5-35B-A3B" \
        --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-1.5-35B-A3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ornith-ai/Ornith-1.5-35B-A3B with Docker Model Runner:
```
docker model run hf.co/ornith-ai/Ornith-1.5-35B-A3B
```

Great work. Bad advertisement.

#17

by Izenflamm - opened about 19 hours ago

Discussion

Izenflamm

about 19 hours ago

•

edited about 19 hours ago

I tested this model for 24 hours and 90M+ tokens. Half on bf16, half on oQ4e.

You did great job, and i mean it. Stability is through the roof, memory is much better than in 1.0. 1.0 named me izenmann (in FILE PATHS!!!) after only 64k tokens. This one holds to the bitter end. Wonderful thinking checkpoints that model makes itself. Model's tenacity is comparable to that of Laguna 2.1 S. I gave it impossible task, far beyond its capabilities, to see where it would break, give up. It didn't. It wasted hours banging it's head against a brick wall, surrounding the bug with tests and mini-harnesses. And the brick wall collapsed. The head is intact. After all, it's just a cyberbone in case of this model :)

The model is dumb as a rock. But it is persistent and stable - more than enough for the most of everyday local agents. But it needs a good harness.

Opencode made it good. DSH made it much better. Hermes is on the list to try.

But guys! You need to advertise smarter. It doesn't beat Claude, it doesn't even come close to the King of the Local AI his magesty Qwen the Dense 27B, even if your trust-me-bro benchmarks says so. But I suppose it would be extremely useful in some cases. SHOW THOSE CASES, not only the benchmarks.

I would try it for Hindsight's "retain", "reflect" and "consolidate" functions. Maybe i'd post-train it with a "onthology thinking" dataset a bit to make it better in isolating entities and facts, then connecting it together. I'm continuing the tests.

fatzin

29 minutes ago

I have the same feeling about this model

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