Instructions to use UnbindAI/Unbind-Flash-V1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use UnbindAI/Unbind-Flash-V1 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 UnbindAI/Unbind-Flash-V1:Q2_K_S
# Run inference directly in the terminal:
llama cli -hf UnbindAI/Unbind-Flash-V1:Q2_K_S

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf UnbindAI/Unbind-Flash-V1:Q2_K_S
# Run inference directly in the terminal:
llama cli -hf UnbindAI/Unbind-Flash-V1:Q2_K_S

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf UnbindAI/Unbind-Flash-V1:Q2_K_S
# Run inference directly in the terminal:
./build/bin/llama cli -hf UnbindAI/Unbind-Flash-V1:Q2_K_S

Use Docker

docker model run hf.co/UnbindAI/Unbind-Flash-V1:Q2_K_S

Llama new
LM Studio
Jan

vLLM

How to use UnbindAI/Unbind-Flash-V1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "UnbindAI/Unbind-Flash-V1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "UnbindAI/Unbind-Flash-V1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/UnbindAI/Unbind-Flash-V1:Q2_K_S

Ollama
How to use UnbindAI/Unbind-Flash-V1 with Ollama:
```
ollama run hf.co/UnbindAI/Unbind-Flash-V1:Q2_K_S
```
Unsloth Desktop

How to use UnbindAI/Unbind-Flash-V1 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf UnbindAI/Unbind-Flash-V1:Q2_K_S

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": "UnbindAI/Unbind-Flash-V1:Q2_K_S"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use UnbindAI/Unbind-Flash-V1 with Docker Model Runner:
```
docker model run hf.co/UnbindAI/Unbind-Flash-V1:Q2_K_S
```

Lemonade

How to use UnbindAI/Unbind-Flash-V1 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull UnbindAI/Unbind-Flash-V1:Q2_K_S

Run and chat with the model

lemonade run user.Unbind-Flash-V1-Q2_K_S

List all available models

lemonade list

Hermes Agent

How to use UnbindAI/Unbind-Flash-V1 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf UnbindAI/Unbind-Flash-V1:Q2_K_S

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 UnbindAI/Unbind-Flash-V1:Q2_K_S

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use UnbindAI/Unbind-Flash-V1 with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf UnbindAI/Unbind-Flash-V1:Q2_K_S

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 "UnbindAI/Unbind-Flash-V1:Q2_K_S" \
  --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"

UnbindAI commited on 10 days ago

Commit

056c08f

verified ·

1 Parent(s): 2635d2f

Complete model card for Unbind Flash V1

Browse files

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -12,14 +12,14 @@ tags:
 # Unbind Flash V1
-Unbind Flash V1 is an Unbind-branded distribution of DeepSeek-V4-Flash in GGUF format. This repository is being prepared and the model weights have not yet been copied.
 ## Model details
 - **Original model:** [DeepSeek-V4-Flash](https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash), by DeepSeek.
 - **Quantized source:** [ggml-org/DeepSeek-V4-Flash-GGUF](https://huggingface.co/ggml-org/DeepSeek-V4-Flash-GGUF), Q2_K_S.
 - **Format:** GGUF, two shards, approximately 98.6 GB in total.
-- **Changes:** Unbind branding and this model card. The model weights are to be copied unchanged from the quantized source; no fine-tuning or independent evaluation has been performed by UnbindAI.
 - **License:** MIT. The original DeepSeek copyright and license notice is reproduced below.
 The source conversion notes that its Q2 models do not use an imatrix calibration. Refer to the source model card for details and limitations.

 # Unbind Flash V1
+Unbind Flash V1 is an Unbind-branded distribution of DeepSeek-V4-Flash in GGUF format. This repository contains the Q2_K_S quantized model weights.
 ## Model details
 - **Original model:** [DeepSeek-V4-Flash](https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash), by DeepSeek.
 - **Quantized source:** [ggml-org/DeepSeek-V4-Flash-GGUF](https://huggingface.co/ggml-org/DeepSeek-V4-Flash-GGUF), Q2_K_S.
 - **Format:** GGUF, two shards, approximately 98.6 GB in total.
+- **Changes:** Unbind branding and this model card. The model weights are unchanged from the quantized source; no fine-tuning or independent evaluation has been performed by UnbindAI.
 - **License:** MIT. The original DeepSeek copyright and license notice is reproduced below.
 The source conversion notes that its Q2 models do not use an imatrix calibration. Refer to the source model card for details and limitations.