Instructions to use datasysdev/ann-sparseattention with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use datasysdev/ann-sparseattention with Transformers:

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("datasysdev/ann-sparseattention", dtype="auto")

llama-cpp-python

How to use datasysdev/ann-sparseattention with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="datasysdev/ann-sparseattention",
	filename="gguf/Qwen3-4B-Instruct-2507-F16-ann-6layer-k128-v2.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use datasysdev/ann-sparseattention 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 datasysdev/ann-sparseattention:F16
# Run inference directly in the terminal:
llama cli -hf datasysdev/ann-sparseattention:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf datasysdev/ann-sparseattention:F16
# Run inference directly in the terminal:
llama cli -hf datasysdev/ann-sparseattention: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 datasysdev/ann-sparseattention:F16
# Run inference directly in the terminal:
./llama-cli -hf datasysdev/ann-sparseattention: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 datasysdev/ann-sparseattention:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf datasysdev/ann-sparseattention:F16

Use Docker

docker model run hf.co/datasysdev/ann-sparseattention:F16

LM Studio
Jan
Ollama
How to use datasysdev/ann-sparseattention with Ollama:
```
ollama run hf.co/datasysdev/ann-sparseattention:F16
```

Unsloth Studio

How to use datasysdev/ann-sparseattention 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 datasysdev/ann-sparseattention 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 datasysdev/ann-sparseattention to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for datasysdev/ann-sparseattention to start chatting

How to use datasysdev/ann-sparseattention with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf datasysdev/ann-sparseattention: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": "datasysdev/ann-sparseattention:F16"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use datasysdev/ann-sparseattention with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf datasysdev/ann-sparseattention: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 datasysdev/ann-sparseattention:F16

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use datasysdev/ann-sparseattention with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf datasysdev/ann-sparseattention:F16

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 "datasysdev/ann-sparseattention:F16" \
  --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 datasysdev/ann-sparseattention with Docker Model Runner:
```
docker model run hf.co/datasysdev/ann-sparseattention:F16
```

Lemonade

How to use datasysdev/ann-sparseattention with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull datasysdev/ann-sparseattention:F16

Run and chat with the model

lemonade run user.ann-sparseattention-F16

List all available models

lemonade list

ann-sparseattention / logs /compare_all32_step1000.log

datasysdev

Update logs/compare_all32_step1000.log

f43f277 verified about 2 months ago

Raw

History Blame Contribute Delete

3.51 kB

	Loading Qwen/Qwen3-4B-Instruct-2507 ...
	Loading weights: 0%\| \| 0/398 [00:00<?, ?it/s] Loading weights: 0%\| \| 1/398 [00:00<01:31, 4.35it/s] Loading weights: 4%\|▍ \| 16/398 [00:00<00:06, 59.27it/s] Loading weights: 9%\|▉ \| 36/398 [00:00<00:03, 108.23it/s] Loading weights: 14%\|█▎ \| 54/398 [00:00<00:02, 130.02it/s] Loading weights: 18%\|█▊ \| 71/398 [00:00<00:02, 139.56it/s] Loading weights: 23%\|██▎ \| 92/398 [00:00<00:01, 157.77it/s] Loading weights: 28%\|██▊ \| 113/398 [00:00<00:01, 169.67it/s] Loading weights: 33%\|███▎ \| 131/398 [00:00<00:01, 168.63it/s] Loading weights: 37%\|███▋ \| 149/398 [00:01<00:01, 170.23it/s] Loading weights: 42%\|████▏ \| 169/398 [00:01<00:01, 175.77it/s] Loading weights: 47%\|████▋ \| 188/398 [00:01<00:01, 179.32it/s] Loading weights: 52%\|█████▏ \| 207/398 [00:01<00:01, 170.67it/s] Loading weights: 57%\|█████▋ \| 225/398 [00:01<00:01, 167.23it/s] Loading weights: 62%\|██████▏ \| 246/398 [00:01<00:00, 177.62it/s] Loading weights: 67%\|██████▋ \| 267/398 [00:01<00:00, 181.73it/s] Loading weights: 72%\|███████▏ \| 286/398 [00:01<00:00, 182.09it/s] Loading weights: 77%\|███████▋ \| 305/398 [00:01<00:00, 180.88it/s] Loading weights: 81%\|████████▏ \| 324/398 [00:02<00:00, 166.37it/s] Loading weights: 87%\|████████▋ \| 345/398 [00:02<00:00, 175.19it/s] Loading weights: 92%\|█████████▏\| 366/398 [00:02<00:00, 179.12it/s] Loading weights: 97%\|█████████▋\| 385/398 [00:02<00:00, 179.11it/s] Loading weights: 100%\|██████████\| 398/398 [00:02<00:00, 161.38it/s]
	Loaded ckpt step 1000 for layers [3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34]
	batch 1/2 done

	========================================================================
	mass@K — fraction of teacher attention captured by retrieval set
	raw_qk : exact top-K over head-mean-aggregated post-RoPE Q,K
	learned: exact top-K over trained search projections (d=128)
	========================================================================

	K method L03 L04 L05 L06 L07 L08 L09 L10 L11 L12 L13 L14 L15 L16 L17 L18 L19 L20 L21 L22 L23 L24 L25 L26 L27 L28 L29 L30 L31 L32 L33 L34 avg
	128 raw_qk 0.939 0.944 0.964 0.956 0.982 0.971 0.959 0.974 0.976 0.961 0.971 0.973 0.968 0.956 0.959 0.965 0.961 0.959 0.966 0.963 0.979 0.971 0.986 0.978 0.978 0.979 0.982 0.988 0.984 0.979 0.977 0.976 0.969
	128 learned 0.924 0.937 0.948 0.939 0.983 0.971 0.977 0.971 0.976 0.970 0.971 0.973 0.973 0.961 0.967 0.972 0.969 0.976 0.980 0.970 0.985 0.979 0.989 0.986 0.983 0.985 0.987 0.987 0.983 0.980 0.967 0.960 0.971

	256 raw_qk 0.986 0.986 0.993 0.990 0.996 0.994 0.992 0.995 0.996 0.991 0.995 0.996 0.995 0.992 0.993 0.995 0.993 0.993 0.994 0.993 0.996 0.994 0.997 0.996 0.995 0.995 0.997 0.998 0.997 0.995 0.995 0.995 0.994
	256 learned 0.977 0.982 0.986 0.981 0.996 0.993 0.995 0.992 0.995 0.993 0.993 0.994 0.995 0.991 0.994 0.995 0.994 0.996 0.997 0.994 0.997 0.996 0.998 0.997 0.997 0.997 0.997 0.997 0.996 0.995 0.991 0.990 0.993

	Learned vs raw mass@K=128: 0.971 / 0.969 = 1.00×

	Wrote /tmp/checkpoints_all32_d128_block_reserve_0_1_2_35/search_step_1000.compare_retrieval.json