Instructions to use giannisan/Laguna-S-2.1-ds4-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 giannisan/Laguna-S-2.1-ds4-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 giannisan/Laguna-S-2.1-ds4-gguf:F32
# Run inference directly in the terminal:
llama cli -hf giannisan/Laguna-S-2.1-ds4-gguf:F32

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf giannisan/Laguna-S-2.1-ds4-gguf:F32
# Run inference directly in the terminal:
llama cli -hf giannisan/Laguna-S-2.1-ds4-gguf:F32

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 giannisan/Laguna-S-2.1-ds4-gguf:F32
# Run inference directly in the terminal:
./llama-cli -hf giannisan/Laguna-S-2.1-ds4-gguf:F32

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 giannisan/Laguna-S-2.1-ds4-gguf:F32
# Run inference directly in the terminal:
./build/bin/llama-cli -hf giannisan/Laguna-S-2.1-ds4-gguf:F32

Use Docker

docker model run hf.co/giannisan/Laguna-S-2.1-ds4-gguf:F32

LM Studio
Jan
Ollama
How to use giannisan/Laguna-S-2.1-ds4-gguf with Ollama:
```
ollama run hf.co/giannisan/Laguna-S-2.1-ds4-gguf:F32
```
Unsloth Desktop

How to use giannisan/Laguna-S-2.1-ds4-gguf with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf giannisan/Laguna-S-2.1-ds4-gguf:F32

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": "giannisan/Laguna-S-2.1-ds4-gguf:F32"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use giannisan/Laguna-S-2.1-ds4-gguf with Docker Model Runner:
```
docker model run hf.co/giannisan/Laguna-S-2.1-ds4-gguf:F32
```

Lemonade

How to use giannisan/Laguna-S-2.1-ds4-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull giannisan/Laguna-S-2.1-ds4-gguf:F32

Run and chat with the model

lemonade run user.Laguna-S-2.1-ds4-gguf-F32

List all available models

lemonade list

Hermes Agent

How to use giannisan/Laguna-S-2.1-ds4-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 giannisan/Laguna-S-2.1-ds4-gguf:F32

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 giannisan/Laguna-S-2.1-ds4-gguf:F32

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use giannisan/Laguna-S-2.1-ds4-gguf with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf giannisan/Laguna-S-2.1-ds4-gguf:F32

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 "giannisan/Laguna-S-2.1-ds4-gguf:F32" \
  --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"

Laguna-S-2.1 (118B) GGUF for pulsar / ds4 (SSD streaming, CUDA)

Mixed-precision GGUF of poolside/Laguna-S-2.1 (118B total / 8B active MoE, 256 experts top-10 + 1 shared) built for SSD-streaming inference engines: routed experts live on disk and stream per token, so the model runs on GPUs that cannot hold it.

Runs on pulsar (Rust + CUDA).

Measured on RTX 5060 Ti 16GB + RTX 4060 Ti 16GB, Ryzen 9900X, Gen5 NVMe (greedy, warm cache):

metric	pulsar
decode	17.3 tok/s
decode + CPU expert lane (`PULSAR_CPU=1`)	22.4 tok/s
prefill (328-token prompt)	~53 tok/s

Per token only 10 of 256 experts per layer are read; attention, the shared expert, and the router stay resident. At this quant ~15.5GB of hot experts stay VRAM-resident (85% cache hit rate on the decode path), which is why the 2-bit build is both smaller and faster than the Q4_K_M edition. Laguna's hybrid attention (one full-window layer every fourth, sliding-window 512 for the rest), per-head output gate, and per-layer-type RoPE (YaRN on full layers, plain on sliding) are all handled by the engine.

Files

file	provenance	recommendation
`Laguna-S-2.1-IQ2XXS-RouterF32-AttnQ8-SExpQ8-OutQ6.gguf`	routed gate/up experts IQ2_XXS with an importance matrix (36 GB)	use this one
`Laguna-S-2.1-Q2K-RouterF32-AttnQ8-SExpQ8-OutQ6.gguf`	routed experts Q2_K, no imatrix (41 GB)	kept for continuity

Both quantize straight from the BF16 checkpoint (235 GB) and keep an identical high-precision layout for everything that steers a token. They differ only in the routed-expert codec: the IQ2_XXS build uses a general-purpose importance matrix so the 2-bit gate/up weights keep the dimensions that matter; the Q2_K build is uniform 2-bit with no calibration. The imatrix file is smaller and closer to the source.

Recipe

The layout targets a streaming expert cache: routed experts must be uniform fixed-size slabs, and everything that steers a token stays high precision. Same design as antirez's GLM-5.2 ds4 build.

Tensors	Type	Why
routed experts, gate/up	IQ2_XXS (imatrix) / Q2_K	streamed from disk per token; uniform slabs
routed experts, down	Q2_K	streamed per token; down is the sensitive one
shared expert + dense layer 0 FFN	Q8_0	resident, paid once
attention q/k/v/output, all layers	Q8_0	resident (source ships F16, converted at quant time)
token embeddings, output head	Q6_K	resident
router (ffn_gate_inp), expert bias, all norms	F32	decision makers stay exact

The router stays F32 because pulsar runs it through an F32 matmul; a quantized ffn_gate_inp is read as raw floats and destroys routing.

Usage

pulsar

git clone https://github.com/giannisanni/pulsar
cd pulsar
CXX=g++-12 cargo build --release -p engine

./target/release/pulsar-cli -m Laguna-S-2.1-IQ2XXS-RouterF32-AttnQ8-SExpQ8-OutQ6.gguf \
    -p "The capital of France is" -n 64
# or the OpenAI-compatible API + web UI: cargo build --release -p serve

Zero-config multi-GPU: pulsar measures each card's PCIe bandwidth at startup, streams experts over the fastest link, and fills spare GPUs with resident hot experts. First run is cold; a .warm sidecar makes every later run start hot. Main knob: PULSAR_CACHE_GB (host expert cache, defaults to measured free RAM minus a reserve).

Downloads last month: 290

GGUF

Model size

118B params

Architecture

laguna

Hardware compatibility

32-bit

Inference Providers NEW

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

Model tree for giannisan/Laguna-S-2.1-ds4-gguf

Base model

poolside/Laguna-S-2.1

Quantized

(94)

this model

Collection including giannisan/Laguna-S-2.1-ds4-gguf

Pulsar

Collection

Rust + CUDA engine that streams giant MoE models from SSD, running models far larger than VRAM. Mixed-precision GGUFs. github.com/giannisanni/pulsar • 3 items • Updated Jul 23