Instructions to use Accio-Lab/occamy-1.0-MLX-3bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Accio-Lab/occamy-1.0-MLX-3bit 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("Accio-Lab/occamy-1.0-MLX-3bit")

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 Accio-Lab/occamy-1.0-MLX-3bit with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "Accio-Lab/occamy-1.0-MLX-3bit"

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "Accio-Lab/occamy-1.0-MLX-3bit"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use Accio-Lab/occamy-1.0-MLX-3bit with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "Accio-Lab/occamy-1.0-MLX-3bit"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "Accio-Lab/occamy-1.0-MLX-3bit"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "Accio-Lab/occamy-1.0-MLX-3bit",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use Accio-Lab/occamy-1.0-MLX-3bit 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 "Accio-Lab/occamy-1.0-MLX-3bit"

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 Accio-Lab/occamy-1.0-MLX-3bit

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use Accio-Lab/occamy-1.0-MLX-3bit with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "Accio-Lab/occamy-1.0-MLX-3bit"

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 "Accio-Lab/occamy-1.0-MLX-3bit" \
  --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"

Occamy 1.0 — MLX 3-bit, group size 64

Candidate release. Mac Metal acceptance is pending. Linux native MLX validation passed; this is not a claim of validated Mac performance or full model quality.

Source: Accio-Lab/occamy-1.0, revision 8f8e0e58a3c9df042be1a3fa2c191fd8047acfb8.

Converted using official mlx 0.32.2 and mlx-lm 0.31.3 native affine quantization. Base precision is 3-bit with group size 64; native router and shared-expert gate modules use 8-bit. This is a text-only qwen3_5_moe export; vision and MTP are not included.

A lossless adapter stacks separate expert weights in numeric expert order before invoking the Qwen3.5 sanitizer exactly once. Quantization and serialization use native APIs. Reload uses the stock loader without an adapter.

Validation: strict stock reload, complete stored floating-value checks, native dequantization of every quantized row, tokenizer/template comparison, and one bounded cached greedy CPU generation with finite logits. The prompt “Compute 2+2. Answer briefly.” returned 4. This limited smoke test is not a quality benchmark.

from mlx_lm import load, generate
model, tokenizer = load("Accio-Lab/occamy-1.0-MLX-3bit")

See SHA256SUMS for artifact hashes and validation_summary.json for validation scope.

Downloads last month: 95

Safetensors

Model size

35B params

Tensor type

U32

BF16

MLX

Hardware compatibility

3-bit

Model tree for Accio-Lab/occamy-1.0-MLX-3bit

Base model

Qwen/Qwen3.6-35B-A3B

Finetuned

Accio-Lab/occamy-1.0

Quantized

(17)

this model