Instructions to use Lythri/Lythri-7B-A4B-Q4_0_8-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use Lythri/Lythri-7B-A4B-Q4_0_8-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Lythri/Lythri-7B-A4B-Q4_0_8-GGUF:Q4_0
# Run inference directly in the terminal:
llama cli -hf Lythri/Lythri-7B-A4B-Q4_0_8-GGUF:Q4_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Lythri/Lythri-7B-A4B-Q4_0_8-GGUF:Q4_0
# Run inference directly in the terminal:
llama cli -hf Lythri/Lythri-7B-A4B-Q4_0_8-GGUF:Q4_0

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 Lythri/Lythri-7B-A4B-Q4_0_8-GGUF:Q4_0
# Run inference directly in the terminal:
./llama-cli -hf Lythri/Lythri-7B-A4B-Q4_0_8-GGUF:Q4_0

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 Lythri/Lythri-7B-A4B-Q4_0_8-GGUF:Q4_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Lythri/Lythri-7B-A4B-Q4_0_8-GGUF:Q4_0

Use Docker

docker model run hf.co/Lythri/Lythri-7B-A4B-Q4_0_8-GGUF:Q4_0

LM Studio
Jan

vLLM

How to use Lythri/Lythri-7B-A4B-Q4_0_8-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Lythri/Lythri-7B-A4B-Q4_0_8-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Lythri/Lythri-7B-A4B-Q4_0_8-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Lythri/Lythri-7B-A4B-Q4_0_8-GGUF:Q4_0

Ollama
How to use Lythri/Lythri-7B-A4B-Q4_0_8-GGUF with Ollama:
```
ollama run hf.co/Lythri/Lythri-7B-A4B-Q4_0_8-GGUF:Q4_0
```
Unsloth Desktop

How to use Lythri/Lythri-7B-A4B-Q4_0_8-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Lythri/Lythri-7B-A4B-Q4_0_8-GGUF:Q4_0

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "Lythri/Lythri-7B-A4B-Q4_0_8-GGUF:Q4_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use Lythri/Lythri-7B-A4B-Q4_0_8-GGUF with Docker Model Runner:
```
docker model run hf.co/Lythri/Lythri-7B-A4B-Q4_0_8-GGUF:Q4_0
```

Lemonade

How to use Lythri/Lythri-7B-A4B-Q4_0_8-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Lythri/Lythri-7B-A4B-Q4_0_8-GGUF:Q4_0

Run and chat with the model

lemonade run user.Lythri-7B-A4B-Q4_0_8-GGUF-Q4_0

List all available models

lemonade list

Hermes Agent

How to use Lythri/Lythri-7B-A4B-Q4_0_8-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Lythri/Lythri-7B-A4B-Q4_0_8-GGUF:Q4_0

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 Lythri/Lythri-7B-A4B-Q4_0_8-GGUF:Q4_0

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use Lythri/Lythri-7B-A4B-Q4_0_8-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Lythri/Lythri-7B-A4B-Q4_0_8-GGUF:Q4_0

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "Lythri/Lythri-7B-A4B-Q4_0_8-GGUF:Q4_0" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Lythri

Lythri-7B-A4B-Q4_0_8-GGUF

This repository contains the Q4_0_8 quantization of Lythri-7B-A4B. Q4_0_8 is a mixed-precision quantization proposed in the Lythri technical report: critical tensors (attention projections, ffn_down, PLE gates) are kept at Q8_0 while the remaining tensors use Q4_0. Both formats support native SIMD dot-product acceleration, delivering Q6_K-level quality with 1.16–1.76× faster prefill on models ≥4B.

For model details, evaluation results and limitations, see the original model card.

Why Q4_0_8

Lythri models are highly sensitive to quantization. Standard 4-bit formats (Q4_0, Q4_K_M) cause noticeable quality loss, while Q6_K preserves quality but is slower due to its multi-level scale structure. Q4_0_8 provides a middle ground:

	Q4_0	Q6_K	Q4_0_8
Quality	✗ Degraded	✓ Good	✓ Near Q6_K
HW Accel	✓ Yes	✗ No	✓ Yes
Speed	Fast	Slow	Fast

Usage

llama.cpp

llama-cli -hf Lythri/Lythri-7B-A4B-Q4_0_8-GGUF -m <file>.gguf --temp 0.7 --top-p 0.9 --top-k 64 --repeat-penalty 1.05

Ollama

ollama run hf.co/Lythri/Lythri-7B-A4B-Q4_0_8-GGUF

LM Studio

Search for Lythri/Lythri-7B-A4B-Q4_0_8-GGUF in the model browser and download.

Q4_0_8 Quantization Guide

Q4_0_8 uses Q4_0 as the base type and overrides critical tensors to Q8_0, achieving Q6_K-level quality with 1.16–1.76× faster prefill via native SIMD dot-product acceleration.

Build llama.cpp

git clone --depth 1 https://github.com/ggerganov/llama.cpp
cd llama.cpp
pip install gguf
cmake -B build -DCMAKE_BUILD_TYPE=Release
cmake --build build -j$(nproc) --target llama-quantize

Convert & Quantize

# Step 1: Convert HF model to F16 GGUF
python3 convert_hf_to_gguf.py /path/to/model \
    --outfile model-F16.gguf --outtype f16

# Step 2: Quantize with Q4_0_8 rules
# Standard Transformer (all architectures):
build/bin/llama-quantize \
    --tensor-type "attn_q=q8_0" \
    --tensor-type "attn_k=q8_0" \
    --tensor-type "attn_v=q8_0" \
    --tensor-type "attn_output=q8_0" \
    --tensor-type "ffn_down=q8_0" \
    model-F16.gguf model-Q4_0_8.gguf q4_0

# For FFN+PLE architectures (e.g. Gemma 4 E2B/E4B), add:
#   --tensor-type "inp_gate=q8_0"
#   --tensor-type "proj=q8_0"

Tensor Assignment Rules

Tensor	Quant	Rationale
`attn_q`, `attn_k`	Q8_0	Softmax sensitivity
`attn_v`, `attn_output`	Q8_0	Residual stream writer
`ffn_down`	Q8_0	Residual stream writer
`ffn_gate`, `ffn_up`	Q4_0	Layer-internal, SiLU-attenuated
`inp_gate`*	Q8_0	Sub-network activation control
`proj`*	Q8_0	AltUp residual projection

*Extended rules for FFN+PLE architectures only.

Citation

@techreport{li2026lythri,
  title       = {Lythri Technical Report},
  author      = {Li, Jiawen},
  year        = {2026},
  institution = {Zenodo},
  doi         = {10.5281/zenodo.23179311},
  url         = {https://doi.org/10.5281/zenodo.23179311}
}