Instructions to use gittensor-model-hub/Qwythos-9B-nsys-SFT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries
PEFT
How to use gittensor-model-hub/Qwythos-9B-nsys-SFT with PEFT:
```
Task type is invalid.
```
Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use gittensor-model-hub/Qwythos-9B-nsys-SFT 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 gittensor-model-hub/Qwythos-9B-nsys-SFT:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf gittensor-model-hub/Qwythos-9B-nsys-SFT:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf gittensor-model-hub/Qwythos-9B-nsys-SFT:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf gittensor-model-hub/Qwythos-9B-nsys-SFT: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 gittensor-model-hub/Qwythos-9B-nsys-SFT:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf gittensor-model-hub/Qwythos-9B-nsys-SFT: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 gittensor-model-hub/Qwythos-9B-nsys-SFT:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf gittensor-model-hub/Qwythos-9B-nsys-SFT:Q4_K_M

Use Docker

docker model run hf.co/gittensor-model-hub/Qwythos-9B-nsys-SFT:Q4_K_M

LM Studio
Jan

vLLM

How to use gittensor-model-hub/Qwythos-9B-nsys-SFT with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "gittensor-model-hub/Qwythos-9B-nsys-SFT"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "gittensor-model-hub/Qwythos-9B-nsys-SFT",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/gittensor-model-hub/Qwythos-9B-nsys-SFT:Q4_K_M

Ollama
How to use gittensor-model-hub/Qwythos-9B-nsys-SFT with Ollama:
```
ollama run hf.co/gittensor-model-hub/Qwythos-9B-nsys-SFT:Q4_K_M
```

Unsloth Studio

How to use gittensor-model-hub/Qwythos-9B-nsys-SFT 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 gittensor-model-hub/Qwythos-9B-nsys-SFT 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 gittensor-model-hub/Qwythos-9B-nsys-SFT to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for gittensor-model-hub/Qwythos-9B-nsys-SFT to start chatting

How to use gittensor-model-hub/Qwythos-9B-nsys-SFT with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf gittensor-model-hub/Qwythos-9B-nsys-SFT: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": "gittensor-model-hub/Qwythos-9B-nsys-SFT:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use gittensor-model-hub/Qwythos-9B-nsys-SFT with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf gittensor-model-hub/Qwythos-9B-nsys-SFT: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 gittensor-model-hub/Qwythos-9B-nsys-SFT:Q4_K_M

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use gittensor-model-hub/Qwythos-9B-nsys-SFT with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf gittensor-model-hub/Qwythos-9B-nsys-SFT: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 "gittensor-model-hub/Qwythos-9B-nsys-SFT: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"

Docker Model Runner
How to use gittensor-model-hub/Qwythos-9B-nsys-SFT with Docker Model Runner:
```
docker model run hf.co/gittensor-model-hub/Qwythos-9B-nsys-SFT:Q4_K_M
```

Lemonade

How to use gittensor-model-hub/Qwythos-9B-nsys-SFT with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull gittensor-model-hub/Qwythos-9B-nsys-SFT:Q4_K_M

Run and chat with the model

lemonade run user.Qwythos-9B-nsys-SFT-Q4_K_M

List all available models

lemonade list

Qwythos-9B-nsys-SFT / eval /summary.json

jared89

Add BF16 eval artifact summary.json

c522c66 verified 11 days ago

Raw

History Blame Contribute Delete

1.88 kB

	{
	"self_taught": {
	"dataset_generator": "local Qwythos-9B via sparkinfer (no external LLM API teacher)",
	"dataset": "gittensor-model-hub/cuda-nsys-training",
	"sft_target": "same Qwythos-9B base (Empero/Qwen pretrained weights + self-generated trajectories)"
	},
	"protocol": {
	"n": 20,
	"source": "data/axolotl/sft_eval.jsonl first 20 rows",
	"turn": "first assistant (messages before first assistant)",
	"temperature": 0.2,
	"max_tokens": 768,
	"metric": "XML <tool_call> parse via spark_agent.parser.parse_tool_calls"
	},
	"base_q4_sparkinfer": {
	"label": "base",
	"n": 20,
	"parse_rate": 0.0,
	"has_tool_call_rate": 0.0,
	"all_valid_tool_rate": 0.0,
	"any_valid_tool_rate": 0.0,
	"has_args_rate": 0.0,
	"gold_parse_rate": 1.0,
	"errors": 0,
	"mean_latency_s": 23.25,
	"name_histogram": {}
	},
	"sft_q4_sparkinfer": {
	"label": "sft",
	"n": 20,
	"parse_rate": 0.3,
	"has_tool_call_rate": 0.3,
	"all_valid_tool_rate": 0.3,
	"any_valid_tool_rate": 0.3,
	"has_args_rate": 0.3,
	"gold_parse_rate": 1.0,
	"errors": 0,
	"mean_latency_s": 11.91,
	"name_histogram": {
	"cudal1_list_tasks": 5,
	"nsys_compare_traces": 1
	}
	},
	"sft_bf16_llamacpp": {
	"label": "sft-bf16-llamacpp-v2",
	"n": 20,
	"parse_rate": 0.85,
	"has_tool_call_rate": 0.85,
	"all_valid_tool_rate": 0.85,
	"any_valid_tool_rate": 0.85,
	"has_args_rate": 0.8,
	"gold_parse_rate": 1.0,
	"errors": 0,
	"mean_latency_s": 1.26,
	"name_histogram": {
	"cudal1_list_tasks": 15,
	"nsys_profile": 1,
	"list_profiling_targets": 1
	}
	},
	"notes": [
	"BF16 eval used llama-server with chat_template_kwargs.enable_thinking=false",
	"Q4 sparkinfer eval did not explicitly disable thinking",
	"BF16 artifact: Qwythos-9B-nsys-SFT-noMTP-BF16.gguf"
	]
	}