Instructions to use dnhkng/RYS-XLarge with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use dnhkng/RYS-XLarge with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="dnhkng/RYS-XLarge") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("dnhkng/RYS-XLarge") model = AutoModelForCausalLM.from_pretrained("dnhkng/RYS-XLarge", device_map="auto") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.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(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use dnhkng/RYS-XLarge with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "dnhkng/RYS-XLarge" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "dnhkng/RYS-XLarge", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/dnhkng/RYS-XLarge
- SGLang
How to use dnhkng/RYS-XLarge 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 "dnhkng/RYS-XLarge" \ --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": "dnhkng/RYS-XLarge", "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 "dnhkng/RYS-XLarge" \ --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": "dnhkng/RYS-XLarge", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use dnhkng/RYS-XLarge with Docker Model Runner:
docker model run hf.co/dnhkng/RYS-XLarge
This is a new kind of model optimization. It is based on a new method for the analysis of the functional role of layers within the transformer stack, and on layer duplication (self-merging) to increase intelligence.
RYS-XLarge at #1 on the HuggingFace Open LLM Leaderboard.
The method is described in full here: LLM Neuroanatomy: How I Topped the LLM Leaderboard Without Changing a Single Weight. If you use this model or the technique in your work, please cite that write-up (BibTeX at the bottom of this card) rather than only the model card.
How it works, briefly
A Transformer stack is not uniform: early layers read the input into an abstract internal representation, late layers write it back out, and the middle layers do the reasoning. Sweeping every contiguous block [i, j) of a pretrained model and re-running it a second time produces a heatmap of which blocks help and which hurt. The blocks that help are clustered in the middle of the stack, and duplicating one of them makes the model measurably better at reasoning benchmarks.
The sweep that produced this model. Each pixel is a full evaluation of one contiguous block (i, j) of Qwen2-72B, re-run a second time: red is improvement, blue is degradation. The green circle marks the configuration that was shipped.
Split by probe, the two tasks disagree about what the stack is doing, which is the interesting part: maths tolerates duplication almost anywhere before layer 60, while EQ shows sharp structure, with good regions walled off by bad ones.
For RYS-XLarge, the chosen configuration was (45, 52): layers 0-51 run, then layers 45-79 run again, so layers 45-51 execute twice. Seven extra layers near the middle of an 80-layer stack, taking the model from 72B to 78B. There was no fine-tuning, no gradient descent and no weight modification of any kind: the extra layers are byte-identical copies of layers the model already had.
The layout is not universal. Every model family puts its duplicable circuits in a different place, so the sweep has to be re-run per architecture - here is Llama-3-70B-Instruct for comparison:
Follow-up posts cover the method on modern architectures and why it works:
- LLM Neuroanatomy II: Modern LLM Hacking and hints of a Universal Language?
- LLM Neuroanatomy III: Why RYS Works - The Language-Agnostic Middle
No Weights were modified in this process!
Model improvement with layer duplication:
| Average | IFEval | BBH | MATH Lvl 5 | GPQA | MUSR | MMLU-PRO | |
|---|---|---|---|---|---|---|---|
| RYS Improvement | 2.61% | -2.05% | 2.51% | 8.16% | 2.58% | 17.72% | 0.31% |
This model is based on MaziyarPanahi/calme-2.1-qwen2-72b, which in turn was tuned from Qwen2-72B. As this method is orthogonal to fine-tuning, the further finetune from MaziyarPanahi now has the top position: https://huggingface.co/MaziyarPanahi/calme-2.4-rys-78b
The method is written up in full in the blog post linked above. At the time of release, all four top models on the leaderboard were based on the RYS method. Special thanks to my wife, for putting up with me coding in the basement for too many evenings and weekends for months!
This research was supported with hardware from the appliedAI Institute, whose goal is to generate and communicate high-quality knowledge about trustworthy AI.
Quickstart
Here is a code snippet with apply_chat_template to show you how to load the tokenizer and model and how to generate content.
from transformers import AutoModelForCausalLM, AutoTokenizer
device = "cuda" # the device to load the model onto
model = AutoModelForCausalLM.from_pretrained(
"dnhkng/RYS-XLarge",
torch_dtype="auto",
device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained("dnhkng/RYS-XLarge")
prompt = "Give me a short introduction to large language model."
messages = [
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(device)
generated_ids = model.generate(
model_inputs.input_ids,
max_new_tokens=512
)
generated_ids = [
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 44.75 |
| IFEval (0-Shot) | 79.96 |
| BBH (3-Shot) | 58.77 |
| MATH Lvl 5 (4-Shot) | 38.97 |
| GPQA (0-shot) | 17.90 |
| MuSR (0-shot) | 23.72 |
| MMLU-PRO (5-shot) | 49.20 |
Citing this work
If this model or the layer-duplication technique is useful in your research, please cite the write-up. It is archived on Zenodo with a DOI, so there is an archival reference if a reviewer objects to citing a blog:
@misc{ng2026rys,
title = {LLM Neuroanatomy: How I Topped the LLM Leaderboard Without Changing a Single Weight},
author = {Ng, David Noel},
year = {2026},
month = {March},
publisher = {Zenodo},
doi = {10.5281/zenodo.24230106},
url = {https://dnhkng.github.io/posts/rys/}
}
DOI: 10.5281/zenodo.24230106 (all versions) Β· ORCID: 0009-0009-3006-4068
ADVERTISING BREAK
Iβm on the hunt for new challenges and a chance to dive into some exciting research opportunities. Oh, and did I mention I just snagged a top spot on the Open LLM leaderboard? π
Profile
Innovation enthusiast, AI strategist, and interdisciplinary-tech nerd β that's me! With over a decade of experience in research and project management, my professional journey has been largely shaped by my passion for artificial intelligence and its potential to transform various industries. With a solid background in artificial intelligence and machine learning, coupled with a knack for innovation and problem-solving (and a healthy dose of curiosity), I'm excited to bring my skills to a new team.
Originally from Australia, where I earned my degrees in Organic Chemistry and Biochemistry, I moved to Germany in 2004. My academic pursuit continued with a Ph.D. in Chemistry at the Max Planck Institute of Biochemistry. Today, I leverage my robust educational background and diverse industry experience to drive AI innovations in a wide range of applications. Hobbies? Lots: I've also built the world's most powerful espresso machine and am working to bring GLaDOS to life.
I'm based out of Munich, Germany, but I would be interested in working remotely for a team with more compute than my 2x 4090s 4x H100s (thanks appliedAI!) π
Reach out via LinkedIn - Dr David Noel Ng
- Downloads last month
- 168
Model tree for dnhkng/RYS-XLarge
Spaces using dnhkng/RYS-XLarge 4
Evaluation results
- strict accuracy on IFEval (0-Shot)Open LLM Leaderboard79.960
- normalized accuracy on BBH (3-Shot)Open LLM Leaderboard58.770
- exact match on MATH Lvl 5 (4-Shot)Open LLM Leaderboard38.970
- acc_norm on GPQA (0-shot)Open LLM Leaderboard17.900
- acc_norm on MuSR (0-shot)Open LLM Leaderboard23.720
- accuracy on MMLU-PRO (5-shot)test set Open LLM Leaderboard49.200



