Instructions to use TitleOS/MindGem27B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TitleOS/MindGem27B with Transformers:

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

pipe = pipeline("image-text-to-text", model="TitleOS/MindGem27B")
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, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("TitleOS/MindGem27B")
model = AutoModelForImageTextToText.from_pretrained("TitleOS/MindGem27B")
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 TitleOS/MindGem27B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "TitleOS/MindGem27B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TitleOS/MindGem27B",
		"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/TitleOS/MindGem27B

SGLang

How to use TitleOS/MindGem27B 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 "TitleOS/MindGem27B" \
    --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": "TitleOS/MindGem27B",
		"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 "TitleOS/MindGem27B" \
        --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": "TitleOS/MindGem27B",
		"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"
						}
					}
				]
			}
		]
	}'

Unsloth Studio

How to use TitleOS/MindGem27B 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 TitleOS/MindGem27B 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 TitleOS/MindGem27B to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for TitleOS/MindGem27B to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="TitleOS/MindGem27B",
    max_seq_length=2048,
)

Docker Model Runner
How to use TitleOS/MindGem27B with Docker Model Runner:
```
docker model run hf.co/TitleOS/MindGem27B
```

MindGem27B

Developed by: TitleOS
License: mpl-2.0
Finetuned from model : unsloth/gemma-3-27b-it-unsloth-bnb-4bit

MindGem 27B is a finetuned QLoRA adapter for unsloth/gemma-3-27b-it-unsloth-bnb-4bit, designed to imbue chain of thought and an internal monologue to Gemma 3 27B. Finetuning was performed on an A100 SXM4, using the Jackrong/Natural-Reasoning-gpt-oss-120B-S1 dataset.

Inference:

--- Prompt ---
What are the most accepted candiates for the unfying theory in physics?

--- Generated Response ---
<reasoning>
We need to answer the question: "What are the most accepted candidates for the unifying theory in physics?" The user likely wants a list of candidate theories, perhaps with commentary on their status and acceptance among physicists.

We need to provide a thorough answer, covering:

- Standard Model: the current successful theory, but not a unification of gravity.
- Grand Unified Theories (GUTs): SU(5), SO(10), E6, etc.
- Superstring theory (including M-theory, bosonic string theory, heterotic string theory, Type IIA, IIB).
- Loop Quantum Gravity (LQG).
- Causal Dynamical Triangulations (CDT).
- Asymptotically Safe Quantum Gravity.
- Twistor theory.
- Noncommutative Geometry.
- Emergent spacetime ideas (e.g., matrix models, holographic principle, AdS/CFT).
- Causal Set theory.
- Other approaches: causal fermion theory, emergent gravity, shape dynamics, etc.

We should also mention "unifying theories" that try to combine quantum mechanics and general relativity, but also potentially unify fundamental forces.

We should discuss the current status: which are most accepted (e.g., string theory has the most research activity and a large community, but no experimental evidence; LQG has a strong community but faces challenges; GUTs have predictions but not observed; asymptotic safety is promising; emergent gravity ideas have some support).

We might also mention "no currently accepted candidate" and the open challenges.

We should give a balanced overview, perhaps with a table summarizing the main features, pros, cons, and community size.

Also discuss experimental constraints: LHC, neutrino masses, proton decay, cosmology (CMB, inflation, dark matter), gravitational wave observations, etc.

We should mention the "unifying" aspect: unification of forces (electromagnetism, weak, strong, gravity), unification of quantum mechanics and general relativity, unification of spacetime and matter.

We could also mention "Theory of Everything" candidates: string theory, M-theory, loop quantum gravity, asymptotic safety, causal dynamical triangulations, emergent gravity, and perhaps "nothing yet" as the most accepted.

We should also discuss "most accepted" as in the most active research, most cited, most funding, most plausible according to theoretical community.

We can categorize:

- Quantum gravity approaches: string theory, loop quantum gravity, causal sets, asymptotic safety,