Instructions to use iluvatarlabs/Valar-27B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use iluvatarlabs/Valar-27B with Transformers:

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

pipe = pipeline("image-text-to-text", model="iluvatarlabs/Valar-27B")
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)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("iluvatarlabs/Valar-27B")
model = AutoModelForMultimodalLM.from_pretrained("iluvatarlabs/Valar-27B", 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=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use iluvatarlabs/Valar-27B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "iluvatarlabs/Valar-27B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "iluvatarlabs/Valar-27B",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/iluvatarlabs/Valar-27B

SGLang

How to use iluvatarlabs/Valar-27B 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 "iluvatarlabs/Valar-27B" \
    --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": "iluvatarlabs/Valar-27B",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "iluvatarlabs/Valar-27B" \
        --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": "iluvatarlabs/Valar-27B",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use iluvatarlabs/Valar-27B with Docker Model Runner:
```
docker model run hf.co/iluvatarlabs/Valar-27B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Valar-27B

Valar-27B is a post-trained derivative of Qwen3.8-27B for scientific investigation and paper-reproduction workflows.

It serves as the investigator: planning experiments, directing an external coding agent, interpreting results, and revising the investigation.

This repository provides the complete BF16 checkpoint in Safetensors format, together with its tokenizer, chat template, and processor configurations. A separate adapter is not required.

Usage

Use a Transformers or vLLM version supporting Qwen3.8-27B. Preserve the supplied tokenizer and chat template.

Agentic scientific workflows require a harness that connects the model to an external coding agent and an execution environment. The weights alone do not provide those tools.

Evaluation

Scientific investigation evaluations used an external coding agent. Results describe the combined investigator-and-coding-agent system.

Attribution

Based on Qwen/Qwen3.8-27B. Model weights were modified through post-training. The upstream Apache-2.0 license is included.

Downloads last month: 18

Safetensors

Model size

28B params

Tensor type

BF16

Model tree for iluvatarlabs/Valar-27B

Base model

Qwen/Qwen3.8-27B

Finetuned

(482)

this model

Quantizations

1 model