Instructions to use RespectMathias/Laguna-XS-2.1-DSpark-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use RespectMathias/Laguna-XS-2.1-DSpark-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 RespectMathias/Laguna-XS-2.1-DSpark-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf RespectMathias/Laguna-XS-2.1-DSpark-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 RespectMathias/Laguna-XS-2.1-DSpark-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf RespectMathias/Laguna-XS-2.1-DSpark-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 RespectMathias/Laguna-XS-2.1-DSpark-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf RespectMathias/Laguna-XS-2.1-DSpark-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 RespectMathias/Laguna-XS-2.1-DSpark-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf RespectMathias/Laguna-XS-2.1-DSpark-GGUF:Q4_K_M

Use Docker

docker model run hf.co/RespectMathias/Laguna-XS-2.1-DSpark-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use RespectMathias/Laguna-XS-2.1-DSpark-GGUF with Ollama:
```
ollama run hf.co/RespectMathias/Laguna-XS-2.1-DSpark-GGUF:Q4_K_M
```

Unsloth Studio

How to use RespectMathias/Laguna-XS-2.1-DSpark-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 RespectMathias/Laguna-XS-2.1-DSpark-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 RespectMathias/Laguna-XS-2.1-DSpark-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for RespectMathias/Laguna-XS-2.1-DSpark-GGUF to start chatting

How to use RespectMathias/Laguna-XS-2.1-DSpark-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RespectMathias/Laguna-XS-2.1-DSpark-GGUF: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": "RespectMathias/Laguna-XS-2.1-DSpark-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use RespectMathias/Laguna-XS-2.1-DSpark-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 RespectMathias/Laguna-XS-2.1-DSpark-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 RespectMathias/Laguna-XS-2.1-DSpark-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use RespectMathias/Laguna-XS-2.1-DSpark-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RespectMathias/Laguna-XS-2.1-DSpark-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 "RespectMathias/Laguna-XS-2.1-DSpark-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"

Docker Model Runner
How to use RespectMathias/Laguna-XS-2.1-DSpark-GGUF with Docker Model Runner:
```
docker model run hf.co/RespectMathias/Laguna-XS-2.1-DSpark-GGUF:Q4_K_M
```

Lemonade

How to use RespectMathias/Laguna-XS-2.1-DSpark-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull RespectMathias/Laguna-XS-2.1-DSpark-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Laguna-XS-2.1-DSpark-GGUF-Q4_K_M

List all available models

lemonade list

Laguna-XS-2.1-DSpark-GGUF

GGUF conversions of RespectMathias/Laguna-XS-2.1-DSpark, a DSpark speculative decoding model for poolside/Laguna-XS-2.1.

This drafter requires Laguna DSpark support from RespectMathias/llama.cpp. It is loaded as a DFlash GGUF because llama.cpp implements DSpark as DFlash plus Markov and confidence heads. Select DSpark proposal behavior with --spec-type draft-dspark.

Files

Quantization	Size
BF16	982.91 MiB
Q8_0	523.88 MiB
Q6_K	405.30 MiB
Q5_K_M	349.84 MiB
Q4_K_M	297.63 MiB
Q3_K_M	253.57 MiB
Q2_K	204.00 MiB

Q8_0 or Q6_K is recommended for draft quality. Lower-bit draft quantization can reduce speculative acceptance enough to offset memory savings.

Usage

llama-server \
  -m Laguna-XS-2.1-MXFP4_MOE.gguf \
  -md Laguna-XS-2.1-DSpark-Q8_0.gguf \
  --spec-type draft-dspark \
  --spec-draft-n-max 15

The checkpoint uses block size 16 and proposes up to 15 draft tokens. It shares tokenizer, token embeddings, and output projection with target model.

Conversion

python convert_hf_to_gguf.py RespectMathias/Laguna-XS-2.1-DSpark \
  --target-model-dir poolside/Laguna-XS-2.1 \
  --outtype bf16 \
  --outfile Laguna-XS-2.1-DSpark-BF16.gguf

Quantizations were generated from BF16 GGUF with llama-quantize.

Status

Runtime validation completed with Laguna-XS-2.1 target and draft-dspark: model loading, fused QKV, auxiliary feature norms, causal draft attention, Laguna SWA RoPE, Markov head, and token acceptance all executed successfully.

Source checkpoint is an experimental small-data training run. Measured acceptance depends on prompt, target quantization, sampling settings, and draft quantization. See source model card for training details and limitations.

License

OpenMDW-1.1. Inherited from poolside/Laguna-XS-2.1.

Downloads last month: 254

GGUF

Model size

0.5B params

Architecture

dflash

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for RespectMathias/Laguna-XS-2.1-DSpark-GGUF

Base model

poolside/Laguna-XS-2.1

Finetuned

RespectMathias/Laguna-XS-2.1-DSpark

Quantized

(1)

this model