Instructions to use john-broadway/Llama-3.2-3B-RYS-21-24-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use john-broadway/Llama-3.2-3B-RYS-21-24-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="john-broadway/Llama-3.2-3B-RYS-21-24-GGUF",
	filename="Llama-3.2-3B-RYS-21-24-Q4_K_M.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use john-broadway/Llama-3.2-3B-RYS-21-24-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf john-broadway/Llama-3.2-3B-RYS-21-24-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf john-broadway/Llama-3.2-3B-RYS-21-24-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf john-broadway/Llama-3.2-3B-RYS-21-24-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf john-broadway/Llama-3.2-3B-RYS-21-24-GGUF:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf john-broadway/Llama-3.2-3B-RYS-21-24-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf john-broadway/Llama-3.2-3B-RYS-21-24-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf john-broadway/Llama-3.2-3B-RYS-21-24-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf john-broadway/Llama-3.2-3B-RYS-21-24-GGUF:Q4_K_M

Use Docker

docker model run hf.co/john-broadway/Llama-3.2-3B-RYS-21-24-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use john-broadway/Llama-3.2-3B-RYS-21-24-GGUF with Ollama:
```
ollama run hf.co/john-broadway/Llama-3.2-3B-RYS-21-24-GGUF:Q4_K_M
```

Unsloth Studio new

How to use john-broadway/Llama-3.2-3B-RYS-21-24-GGUF 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 john-broadway/Llama-3.2-3B-RYS-21-24-GGUF 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 john-broadway/Llama-3.2-3B-RYS-21-24-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for john-broadway/Llama-3.2-3B-RYS-21-24-GGUF to start chatting

Pi new

How to use john-broadway/Llama-3.2-3B-RYS-21-24-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf john-broadway/Llama-3.2-3B-RYS-21-24-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "john-broadway/Llama-3.2-3B-RYS-21-24-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use john-broadway/Llama-3.2-3B-RYS-21-24-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf john-broadway/Llama-3.2-3B-RYS-21-24-GGUF:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default john-broadway/Llama-3.2-3B-RYS-21-24-GGUF:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use john-broadway/Llama-3.2-3B-RYS-21-24-GGUF with Docker Model Runner:
```
docker model run hf.co/john-broadway/Llama-3.2-3B-RYS-21-24-GGUF:Q4_K_M
```

Lemonade

How to use john-broadway/Llama-3.2-3B-RYS-21-24-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull john-broadway/Llama-3.2-3B-RYS-21-24-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Llama-3.2-3B-RYS-21-24-GGUF-Q4_K_M

List all available models

lemonade list

Llama-3.2-3B-RYS-21-24

Llama-3.2-3B-Instruct with layers 21-23 duplicated. A late-stack math circuit runs twice on every forward pass.

28 base layers → 31 after duplication. No training, no merging, no weight changes.

Math 0.470 → 0.6132 (+14.32 — the biggest math lift in the v2 corpus). Reasoning 88.24% → 82.36% (−5.88). EQ 84.38 → 84.11 (−0.27).

Results

Metric	Baseline	RYS (21,24)	Delta
Math	0.470	0.6132	+14.32
EQ	84.38	84.11	−0.27
Reasoning	88.24%	82.36%	−5.88

The math amplifier. Llama-3.2-3B-Instruct has the second-highest baseline reasoning in the v2 corpus (88.24%) — near-ceiling, so RYS has little reasoning room to lift. But the same train-free intervention applied to the late-stack block (21,24) produces the biggest math lift anywhere in the corpus (+14.32 absolute, ~30% relative). Math and reasoning circuits sit at different depths in this model; the math one has headroom.

Pick this when math throughput matters and reasoning is already strong enough. The within-family contrast is the loudest in the corpus: sibling Llama-3.2-1B-RYS-10-13-GGUF lifts reasoning from 0% → 64.71% on the same block-duplication mechanism applied to a different depth.

Usage

llama-server -m Llama-3.2-3B-RYS-21-24-Q4_K_M.gguf -ngl 99

Full sweep data

54 configurations tested. (21,24) block-3 is the best-combined pick (math-optimal). Full per-config sweep + cross-architecture analysis: v2 dataset.

Part of the RYS Sovereign Collection v2.

Where this sits in the Sovereign Collection

v1 — Qwen2.5 cross-scale + Qwen3-32B headline crossover. 5 model repos.

v2 — cross-architecture corpus. 21 model variants across 10 architecture families. Inverse correlation (r = −0.726): weak baselines lift more, in their weakest dimension. The Llama-3.2 family alone (1B + 3B) spans the entire baseline-vs-magnitude curve in the v2 corpus. 13 deployable RYS-applied weight repos covering every non-zero-lift variant.

Within-family sibling: john-broadway/Llama-3.2-1B-RYS-10-13-GGUF — the 0%→64.71% reasoning unlock at the weak-baseline end.

Credit

John Broadway, with collaboration from Claude (Opus 4.6 in April 2026 sweep generation and build pipeline; Opus 4.7 in May 2026 cross-architecture analysis and publication). Original RYS method by David Ng on Qwen2-72B; sweep + probe toolkit by alainnothere.

Downloads last month: 139

GGUF

Model size

4B params

Architecture

llama

Hardware compatibility

4-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for john-broadway/Llama-3.2-3B-RYS-21-24-GGUF

Base model

meta-llama/Llama-3.2-3B-Instruct

Quantized

(453)

this model

Collection including john-broadway/Llama-3.2-3B-RYS-21-24-GGUF

RYS Sovereign Collection v2

Collection

Qwen3-family cohort (0.6B/1.7B/8B) + cross-architecture sweep corpus (21 variants, 10 families). v1 collection: Qwen2.5 cross-scale + Qwen3-32B. • 18 items • Updated 3 days ago