Instructions to use prithivMLmods/XHToken-Spark-X2.6-4B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use prithivMLmods/XHToken-Spark-X2.6-4B-GGUF with Transformers:

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

pipe = pipeline("text-generation", model="prithivMLmods/XHToken-Spark-X2.6-4B-GGUF")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("prithivMLmods/XHToken-Spark-X2.6-4B-GGUF", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use prithivMLmods/XHToken-Spark-X2.6-4B-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 prithivMLmods/XHToken-Spark-X2.6-4B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf prithivMLmods/XHToken-Spark-X2.6-4B-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 prithivMLmods/XHToken-Spark-X2.6-4B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf prithivMLmods/XHToken-Spark-X2.6-4B-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 prithivMLmods/XHToken-Spark-X2.6-4B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf prithivMLmods/XHToken-Spark-X2.6-4B-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 prithivMLmods/XHToken-Spark-X2.6-4B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf prithivMLmods/XHToken-Spark-X2.6-4B-GGUF:Q4_K_M

Use Docker

docker model run hf.co/prithivMLmods/XHToken-Spark-X2.6-4B-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use prithivMLmods/XHToken-Spark-X2.6-4B-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "prithivMLmods/XHToken-Spark-X2.6-4B-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": "prithivMLmods/XHToken-Spark-X2.6-4B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/prithivMLmods/XHToken-Spark-X2.6-4B-GGUF:Q4_K_M

SGLang

How to use prithivMLmods/XHToken-Spark-X2.6-4B-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 "prithivMLmods/XHToken-Spark-X2.6-4B-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": "prithivMLmods/XHToken-Spark-X2.6-4B-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 "prithivMLmods/XHToken-Spark-X2.6-4B-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": "prithivMLmods/XHToken-Spark-X2.6-4B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use prithivMLmods/XHToken-Spark-X2.6-4B-GGUF with Ollama:
```
ollama run hf.co/prithivMLmods/XHToken-Spark-X2.6-4B-GGUF:Q4_K_M
```
Unsloth Desktop

How to use prithivMLmods/XHToken-Spark-X2.6-4B-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf prithivMLmods/XHToken-Spark-X2.6-4B-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": "prithivMLmods/XHToken-Spark-X2.6-4B-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use prithivMLmods/XHToken-Spark-X2.6-4B-GGUF with Docker Model Runner:
```
docker model run hf.co/prithivMLmods/XHToken-Spark-X2.6-4B-GGUF:Q4_K_M
```

Lemonade

How to use prithivMLmods/XHToken-Spark-X2.6-4B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull prithivMLmods/XHToken-Spark-X2.6-4B-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.XHToken-Spark-X2.6-4B-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use prithivMLmods/XHToken-Spark-X2.6-4B-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 prithivMLmods/XHToken-Spark-X2.6-4B-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 prithivMLmods/XHToken-Spark-X2.6-4B-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use prithivMLmods/XHToken-Spark-X2.6-4B-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf prithivMLmods/XHToken-Spark-X2.6-4B-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 "prithivMLmods/XHToken-Spark-X2.6-4B-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"

XHToken-Spark-X2.6-4B-GGUF

Spark-X2.6-4B is a compact 4-billion-parameter open-weights language model developed by the SparkLLM Team (XHToken/Spark-X2.6-4B, Apache 2.0), post-trained from XHToken/Spark-X2.6-4B-Base for resource-conscious, on-device agentic deployment. Retaining the Spark-X2.5-4B architecture while introducing an improved post-training recipe, the model delivers substantial gains in software engineering, multi-step planning, complex tool use, math, multilingual understanding, and ultra-long-context processing (scaling up to a 1M-token context window). It incorporates a controllable thinking budget with three configurable reasoning levels—none, low, and high—allowing applications to dynamically trade off reasoning depth, latency, and inference cost. Pairing edge inference efficiency that matches or exceeds comparable models like Qwen3.5-4B and Gemma-E4B with native deployment support across Transformers, SGLang, vLLM, MLX, llama.cpp, and Unsloth, Spark-X2.6-4B sets a state-of-the-art standard among 4B-class models and rivals or surpasses larger 9B-parameter models across agentic, coding, and general benchmarks.

Model Files

File Name	Quant Type	File Size	File Link	Description
Spark-X2.6-4B.BF16.gguf	BF16	8.23 GB	Link	Full BF16 weights. Highest quality, largest file size.
Spark-X2.6-4B.Q3_K_L.gguf	Q3_K_L	2.39 GB	Link	Lower quality but usable, good for low RAM availability.
Spark-X2.6-4B.Q3_K_M.gguf	Q3_K_M	2.17 GB	Link	Low quality.
Spark-X2.6-4B.Q4_K_M.gguf	Q4_K_M	2.6 GB	Link	Good quality, default size for most use cases, recommended.
Spark-X2.6-4B.Q4_K_S.gguf	Q4_K_S	2.43 GB	Link	Slightly lower quality with more space savings, recommended.
Spark-X2.6-4B.Q5_K_M.gguf	Q5_K_M	2.98 GB	Link	High quality, recommended.
Spark-X2.6-4B.Q5_K_S.gguf	Q5_K_S	2.88 GB	Link	High quality, recommended.
Spark-X2.6-4B.Q6_K.gguf	Q6_K	3.38 GB	Link	Very high quality, near perfect, recommended.
Spark-X2.6-4B.Q8_0.gguf	Q8_0	4.38 GB	Link	Extremely high quality, generally unneeded but max available quant.