Instructions to use saishshinde15/Clyrai_Vortex_Reasoning_GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use saishshinde15/Clyrai_Vortex_Reasoning_GGUF with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="saishshinde15/Clyrai_Vortex_Reasoning_GGUF")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("saishshinde15/Clyrai_Vortex_Reasoning_GGUF", dtype="auto")

llama-cpp-python

How to use saishshinde15/Clyrai_Vortex_Reasoning_GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="saishshinde15/Clyrai_Vortex_Reasoning_GGUF",
	filename="unsloth.F16.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use saishshinde15/Clyrai_Vortex_Reasoning_GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf saishshinde15/Clyrai_Vortex_Reasoning_GGUF:F16
# Run inference directly in the terminal:
llama-cli -hf saishshinde15/Clyrai_Vortex_Reasoning_GGUF:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf saishshinde15/Clyrai_Vortex_Reasoning_GGUF:F16
# Run inference directly in the terminal:
llama-cli -hf saishshinde15/Clyrai_Vortex_Reasoning_GGUF:F16

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 saishshinde15/Clyrai_Vortex_Reasoning_GGUF:F16
# Run inference directly in the terminal:
./llama-cli -hf saishshinde15/Clyrai_Vortex_Reasoning_GGUF:F16

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 saishshinde15/Clyrai_Vortex_Reasoning_GGUF:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf saishshinde15/Clyrai_Vortex_Reasoning_GGUF:F16

Use Docker

docker model run hf.co/saishshinde15/Clyrai_Vortex_Reasoning_GGUF:F16

LM Studio
Jan

vLLM

How to use saishshinde15/Clyrai_Vortex_Reasoning_GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "saishshinde15/Clyrai_Vortex_Reasoning_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": "saishshinde15/Clyrai_Vortex_Reasoning_GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/saishshinde15/Clyrai_Vortex_Reasoning_GGUF:F16

SGLang

How to use saishshinde15/Clyrai_Vortex_Reasoning_GGUF 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 "saishshinde15/Clyrai_Vortex_Reasoning_GGUF" \
    --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": "saishshinde15/Clyrai_Vortex_Reasoning_GGUF",
		"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 "saishshinde15/Clyrai_Vortex_Reasoning_GGUF" \
        --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": "saishshinde15/Clyrai_Vortex_Reasoning_GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use saishshinde15/Clyrai_Vortex_Reasoning_GGUF with Ollama:
```
ollama run hf.co/saishshinde15/Clyrai_Vortex_Reasoning_GGUF:F16
```

Unsloth Studio new

How to use saishshinde15/Clyrai_Vortex_Reasoning_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 saishshinde15/Clyrai_Vortex_Reasoning_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 saishshinde15/Clyrai_Vortex_Reasoning_GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for saishshinde15/Clyrai_Vortex_Reasoning_GGUF to start chatting

Pi new

How to use saishshinde15/Clyrai_Vortex_Reasoning_GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf saishshinde15/Clyrai_Vortex_Reasoning_GGUF:F16

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": "saishshinde15/Clyrai_Vortex_Reasoning_GGUF:F16"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use saishshinde15/Clyrai_Vortex_Reasoning_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 saishshinde15/Clyrai_Vortex_Reasoning_GGUF:F16

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 saishshinde15/Clyrai_Vortex_Reasoning_GGUF:F16

Run Hermes

hermes

Docker Model Runner
How to use saishshinde15/Clyrai_Vortex_Reasoning_GGUF with Docker Model Runner:
```
docker model run hf.co/saishshinde15/Clyrai_Vortex_Reasoning_GGUF:F16
```

Lemonade

How to use saishshinde15/Clyrai_Vortex_Reasoning_GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull saishshinde15/Clyrai_Vortex_Reasoning_GGUF:F16

Run and chat with the model

lemonade run user.Clyrai_Vortex_Reasoning_GGUF-F16

List all available models

lemonade list

Clyrai Vortex Reasoning (GGUF)

Model Name: saishshinde15/Clyrai_Vortex_Reasoning_GGUF
Developed by: clyrai
License: Apache 2.0
Fine-tuned from: Clyrai_Base_Reasoning
Available in: 16-bit and 4-bit GGUF formats

Overview

TethysAI Vortex Reasoning is an experimental model designed to replicate the advanced reasoning abilities of Clyrai_Base_Reasoning, which was originally enhanced using GRPO. Instead of GRPO, this model was fine-tuned with high-quality structured data using high-end Supervised Fine-Tuning (SFT) to replicate the step-by-step thinking and self-questioning mechanisms seen in models like DeepSeek-R1.

This model has been optimized for efficient inference in GGUF format, allowing for deployment on CPU-based systems and lightweight edge devices without sacrificing reasoning capabilities.

Why This Model Stands Out

🔹 Advanced Self-Reasoning:

The model questions itself internally before arriving at an answer.
Similar to DeepSeek-R1, it follows a structured reasoning process.
Uses and tokens internally, though they may not always be explicitly visible in responses.

🔹 No GRPO, Only High-End SFT:

Instead of GRPO, the model learns structured reasoning directly from fine-tuned data.
Demonstrates logical breakdowns, multi-step problem-solving, and contextual understanding.
Achieves results comparable to the base model without reinforcement learning.

🔹 Optimized for GGUF Inference:

Available in both 16-bit and 4-bit GGUF, enabling fast and memory-efficient execution on CPUs.
Ideal for on-device deployment, including edge computing, embedded AI, and AI assistants.

Usage

Use the below prompt for the best results:

You are an advanced AI assistant. Provide answers in a clear, step-by-step manner.

Downloads last month: 41

GGUF

Model size

3B params

Architecture

qwen2

Hardware compatibility

4-bit

16-bit

Model tree for saishshinde15/Clyrai_Vortex_Reasoning_GGUF

Base model

Qwen/Qwen2.5-3B

Finetuned

Qwen/Qwen2.5-3B-Instruct

Finetuned

saishshinde15/Clyrai_Base_Reasoning

Quantized

(3)

this model

Collection including saishshinde15/Clyrai_Vortex_Reasoning_GGUF

Experimental Model

Collection

Tried to mimic GRPO but using SFT • 2 items • Updated Mar 3, 2025