Instructions to use empero-ai/Qwythos-9B-v2-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 empero-ai/Qwythos-9B-v2-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 empero-ai/Qwythos-9B-v2-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf empero-ai/Qwythos-9B-v2-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf empero-ai/Qwythos-9B-v2-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf empero-ai/Qwythos-9B-v2-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 empero-ai/Qwythos-9B-v2-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf empero-ai/Qwythos-9B-v2-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 empero-ai/Qwythos-9B-v2-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf empero-ai/Qwythos-9B-v2-GGUF:Q4_K_M

Use Docker

docker model run hf.co/empero-ai/Qwythos-9B-v2-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use empero-ai/Qwythos-9B-v2-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "empero-ai/Qwythos-9B-v2-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": "empero-ai/Qwythos-9B-v2-GGUF",
		"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/empero-ai/Qwythos-9B-v2-GGUF:Q4_K_M

Ollama
How to use empero-ai/Qwythos-9B-v2-GGUF with Ollama:
```
ollama run hf.co/empero-ai/Qwythos-9B-v2-GGUF:Q4_K_M
```
Unsloth Desktop

How to use empero-ai/Qwythos-9B-v2-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf empero-ai/Qwythos-9B-v2-GGUF:Q4_K_M

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": "empero-ai/Qwythos-9B-v2-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use empero-ai/Qwythos-9B-v2-GGUF with Docker Model Runner:
```
docker model run hf.co/empero-ai/Qwythos-9B-v2-GGUF:Q4_K_M
```

Lemonade

How to use empero-ai/Qwythos-9B-v2-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull empero-ai/Qwythos-9B-v2-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Qwythos-9B-v2-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use empero-ai/Qwythos-9B-v2-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 empero-ai/Qwythos-9B-v2-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 empero-ai/Qwythos-9B-v2-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use empero-ai/Qwythos-9B-v2-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf empero-ai/Qwythos-9B-v2-GGUF:Q4_K_M

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 "empero-ai/Qwythos-9B-v2-GGUF:Q4_K_M" \
  --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"

Garbled response from Q8_0

by cnsiva - opened Jul 12

Discussion

cnsiva

Jul 12

I've tested Qwythos-9B-v2-Q8_0.gguf and I'm getting garbled responses. My GPU is an RTX 4070 Ti Super. Here is my configuration. The same configuration works with Q6_K

llama-server
--model ~/models/empero-ai/Qwythos-9B-v2-GGUF/Qwythos-9B-v2-Q8_0.gguf
--alias qwythos-9b-pro
--path ~/github-sources/llama.cpp/build/tools/ui/dist
--mmproj ~/models/empero-ai/Qwythos-9B-v2-GGUF/mmproj-Qwythos-9B-v2-BF16.gguf
--n-gpu-layers 999
--n-cpu-moe 0
--cache-type-k q8_0
--cache-type-v q8_0
--parallel 4
--cache-ram 24576
--cache-reuse 256
--kv-unified
--cache-idle-slots
--context-shift
--slot-save-path ~/.cache/llama-slots/qwythos-9b
--cont-batching
--threads 8
--threads-batch 8
--threads-http 4
--prio 2
--prio-batch 1
--numa isolate
--flash-attn on
--ctx-size 262144
--batch-size 4096
--ubatch-size 512
--reasoning auto
--reasoning-format auto
--reasoning-budget -1
--reasoning-budget-message " [Logic Finalized] "
--jinja
--temp 0.6
--top-p 0.95
--top-k 20
--host 0.0.0.0
--port 8080
--log-disable
--metrics

hgeist

Jul 13

Try without kv cache quanization. q35 architecture is very sensitive to it.

cnsiva

Jul 13

It works fine with the MTP version.

empero-ai

Owner Jul 13

Hey @cnsiva

We have tested the model locally and did not find any issues with the Q8 variant.

We would recommend you to remove the cache quantization and the reasoning budget

cnsiva

Jul 14

The MTP variant works seamlessly with Q8_0. However, I’ve noticed that the file/picture input doesn’t function correctly ( —mmproj ~/models/empero-ai/Qwythos-9B-v2-GGUF/mmproj-Qwythos-9B-v2-BF16.gguf).
On the other hand, it performs exceptionally well without the mmproj, achieving a remarkable speed of 120+ t/s, which is truly impressive.

Here is my llama.cpp service file for this model without ( —mmproj ~/models/empero-ai/Qwythos-9B-v2-GGUF/mmproj-Qwythos-9B-v2-BF16.gguf).

[Unit]
Description=Llama.cpp Server - Config: qwythos-9b.conf
After=network.target nss-lookup.target
Wants=nvidia-suspend.service nvidia-hibernate.service nvidia-resume.service

[Service]
Type=simple
User=siva
CPUAffinity=0-7,16-23
LimitMEMLOCK=infinity

Environment=GGML_CUDA_REGISTER_HOST=1

ExecStart=~/github-sources/llama.cpp/build/bin/llama-server
--model ~/models/empero-ai/Qwythos-9B-v2-GGUF/Qwythos-9B-v2-MTP-Q6_K.gguf
--alias qwythos-9b-pro
--path ~/github-sources/llama.cpp/build/tools/ui/dist
--n-gpu-layers 999
--n-cpu-moe 0
--cache-type-k q8_0
--cache-type-v q8_0
--parallel 6
--cache-ram 24576
--cache-reuse 256
--kv-unified
--cache-idle-slots
--slot-save-path /home/siva/.cache/llama-slots/qwythos-9b
--cont-batching
--threads 8
--threads-batch 8
--threads-http 4
--prio 2
--prio-batch 1
--numa isolate
--flash-attn on
--ctx-size 262144
--batch-size 4096
--ubatch-size 512
--reasoning auto
--reasoning-format auto
--reasoning-budget -1
--reasoning-budget-message "[Logic Finalized]"
--jinja
--temp 0.6
--top-p 0.95
--top-k 20
--host 0.0.0.0
--port 8080
--log-disable
--spec-type draft-mtp
--spec-draft-n-max 3
--image-min-tokens 1024
--metrics

Restart=always
RestartSec=5
StandardOutput=journal
StandardError=journal
SyslogIdentifier=llama-server

[Install]
WantedBy=multi-user.target

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment