Instructions to use OsaurusAI/Laguna-XS.2-mxfp4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OsaurusAI/Laguna-XS.2-mxfp4 with MLX:

# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm

# Generate text with mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("OsaurusAI/Laguna-XS.2-mxfp4")

prompt = "Write a story about Einstein"
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True
)

text = generate(model, tokenizer, prompt=prompt, verbose=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

How to use OsaurusAI/Laguna-XS.2-mxfp4 with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "OsaurusAI/Laguna-XS.2-mxfp4"

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "OsaurusAI/Laguna-XS.2-mxfp4"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use OsaurusAI/Laguna-XS.2-mxfp4 with Hermes Agent:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "OsaurusAI/Laguna-XS.2-mxfp4"

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 OsaurusAI/Laguna-XS.2-mxfp4

Run Hermes

hermes

OpenClaw new

How to use OsaurusAI/Laguna-XS.2-mxfp4 with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "OsaurusAI/Laguna-XS.2-mxfp4"

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 "OsaurusAI/Laguna-XS.2-mxfp4" \
  --custom-provider-id mlx-lm \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

MLX LM

How to use OsaurusAI/Laguna-XS.2-mxfp4 with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "OsaurusAI/Laguna-XS.2-mxfp4"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "OsaurusAI/Laguna-XS.2-mxfp4"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "OsaurusAI/Laguna-XS.2-mxfp4",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

OsaurusAI/Laguna-XS.2-mxfp4

Quantized poolside Laguna-XS.2 for Apple Silicon (MLX) — agentic-coding 33B-active-3B Mixture-of-Experts.


Source	poolside/Laguna-XS.2
Architecture	`laguna` (40 layers, 256 routed experts top-8 + 1 shared, hybrid SWA+full attention)
Quant format	MXFP4 (mlx 4-bit affine, group_size=32)
Bundle size on disk	20.93 GB (21 safetensors shards)
License	Apache-2.0 (inherits from upstream)
Modalities	Text in / text out (no vision, no audio, no video)

What's quantized

All routed-expert linears (3D stacked), attention, dense layer-0 MLP, shared-expert, embed_tokens → mlx 4-bit affine (bits=4 group_size=32)
lm_head + all RMSNorms + router gate + e_score_correction_bias → fp16 passthrough

Architecture notes (preserved verbatim from upstream)

40 layers; per-layer attention head count alternates 48 (full-attn) / 64 (SWA) with shared 8 KV heads (GQA)
1:3 ratio of full-attn ↔ sliding-window-attention (window = 512), explicit layer_types list
Dual RoPE: full-attn = YaRN (base 500K, factor 32, original 4096, β_fast 64, β_slow 1, partial_rotary 0.5); SWA = default (base 10K, full rotary)
256 routed experts (top-8) + 1 shared expert; sigmoid + per-head gating (g_proj); q_norm/k_norm in attention
131k context window
Layer 0 dense MLP; layers 1-39 sparse MoE

Run on Apple Silicon

pip install mlx safetensors transformers
python -m jang_tools.laguna.runtime \
    --src ~/.mlxstudio/models/OsaurusAI/Laguna-XS.2-mxfp4 \
    --prompt "def fibonacci(n):" --max-new 64

The runtime auto-detects weight_format (mxtq / mxfp4 / bf16) and loads the matching path (jang_tools/laguna/weight_loader_bf16.py).

Build

Reproduce locally from the bf16 source:

python -m jang_tools.convert_laguna_mxfp4 \
    ~/.mlxstudio/models/_sources/Laguna-XS.2 \
    ~/.mlxstudio/models/OsaurusAI/Laguna-XS.2-mxfp4

Credits

Quantized by Jinho Jang (eric@osaurus.ai). MLX-native pipeline, runs on M-series Macs.

Downloads last month: 44

Safetensors

Model size

6B params

Tensor type

U32

F16

MLX

Hardware compatibility

Quantized

Model tree for OsaurusAI/Laguna-XS.2-mxfp4

Base model

poolside/Laguna-XS.2

Finetuned

(24)

this model