Instructions to use imaadd05/Mellum2.1-12B-A2.5B-Thinking-mlx-6bit-g64 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use imaadd05/Mellum2.1-12B-A2.5B-Thinking-mlx-6bit-g64 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("imaadd05/Mellum2.1-12B-A2.5B-Thinking-mlx-6bit-g64")

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 imaadd05/Mellum2.1-12B-A2.5B-Thinking-mlx-6bit-g64 with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "imaadd05/Mellum2.1-12B-A2.5B-Thinking-mlx-6bit-g64"

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": "imaadd05/Mellum2.1-12B-A2.5B-Thinking-mlx-6bit-g64"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use imaadd05/Mellum2.1-12B-A2.5B-Thinking-mlx-6bit-g64 with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "imaadd05/Mellum2.1-12B-A2.5B-Thinking-mlx-6bit-g64"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "imaadd05/Mellum2.1-12B-A2.5B-Thinking-mlx-6bit-g64"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "imaadd05/Mellum2.1-12B-A2.5B-Thinking-mlx-6bit-g64",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use imaadd05/Mellum2.1-12B-A2.5B-Thinking-mlx-6bit-g64 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 "imaadd05/Mellum2.1-12B-A2.5B-Thinking-mlx-6bit-g64"

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 imaadd05/Mellum2.1-12B-A2.5B-Thinking-mlx-6bit-g64

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use imaadd05/Mellum2.1-12B-A2.5B-Thinking-mlx-6bit-g64 with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "imaadd05/Mellum2.1-12B-A2.5B-Thinking-mlx-6bit-g64"

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 "imaadd05/Mellum2.1-12B-A2.5B-Thinking-mlx-6bit-g64" \
  --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"

Mellum2.1-12B-A2.5B-Thinking-mlx-6bit-g64 / conversion.json

imaadd05

Stage checksum-validated 6-bit candidate pending native Mac release gates

9104747 verified about 19 hours ago

Raw History Blame Contribute Delete

2.9 kB

	{
	"source": "JetBrains/Mellum2.1-12B-A2.5B-Thinking",
	"source_revision": "92ddae9fc7665e9f801d141d2e5a6b2caf2460c4",
	"created_utc": "2026-10-09T08:32:06.014227+00:00",
	"quantization": {
	"mode": "affine",
	"bits": 6,
	"group_size": 64,
	"router_bits": 8,
	"non_quantized_dtype": "bfloat16"
	},
	"device": "gpu",
	"platform": "Linux-7.0.0-38-generic-x86_64-with-glibc2.39",
	"python": "3.12.3",
	"packages": {
	"mlx": "0.32.3",
	"mlx-lm": "0.32.0",
	"transformers": "5.19.0",
	"huggingface-hub": "1.33.0",
	"safetensors": "0.8.0",
	"numpy": "2.5.3"
	},
	"source_sha256": {
	"README.md": "3da5f8b9862c7b3560e631d778e2402f222d87f1a332c34696dfa399608824a7",
	"chat_template.jinja": "4593a34d52f3364ac13ce57f4bea5688924e8942da6df8201e411db17e729f48",
	"config.json": "f43d018246094dee6dca455b3072766b6acbefd1ac8e7fe664df34710606a541",
	"generation_config.json": "a9dc104ca398a2ef376d3f2fbb03037d99950ddb2db0cd041b98398cf1fc8d7c",
	"model-00001-of-00005.safetensors": "f5ccf8012c7d44fd2598325bf2482b8fc21792ef888de68f9ca8c9053c2e022f",
	"model-00002-of-00005.safetensors": "266beea6f34dea9c86414d8bf8cb35c0f4a53330731d99ed41ec666390758e39",
	"model-00003-of-00005.safetensors": "da9a04fb6ea121f71e30920c7657b889ca0078aea57d878e5cb768ebb3e4483a",
	"model-00004-of-00005.safetensors": "32f27ab7f38ff5d98ac3d50dabb57ab8a55a8aa552febd42c2a5b01f4116262a",
	"model-00005-of-00005.safetensors": "d8bbb9b829cb3279e23f59af55bf68ec00cc8ad30923c227d8b13c7d831e6b3b",
	"model.safetensors.index.json": "a6ad930e5bb35ea92d5b734c4df4649bfcfdcd899f64b0a83efe40fe0ab9b60e",
	"special_tokens_map.json": "20e1ead0b0986b55989aa8116bafe63efcbae0daf8ed047970cd28b36aada4d9",
	"tokenizer.json": "58548a346eb073e5132bf7d8ad17dc6971bca36ade378ca4d2bfbc49bf60da2a",
	"tokenizer_config.json": "e2950960d2a3040a1938e6357c8c3628d531133c464286a7869b2e0504fcc59e"
	},
	"artifact_sha256": {
	"chat_template.jinja": "4593a34d52f3364ac13ce57f4bea5688924e8942da6df8201e411db17e729f48",
	"config.json": "0254d0f058f4ccd9f34d3e0b1e85dd425af417da7569a3672b62d692798aaa3e",
	"generation_config.json": "a9dc104ca398a2ef376d3f2fbb03037d99950ddb2db0cd041b98398cf1fc8d7c",
	"model-00001-of-00002.safetensors": "ff781996bc063c39d4dd01bc86a664522c06277412cfba8b48063a189fdd33e0",
	"model-00002-of-00002.safetensors": "d92c83c9a08ce5d9f6a493d3f851dfbb15cf12a3a30dc856674eecf9e1ea283a",
	"model.safetensors.index.json": "aef292869670c96d74e8ae0fd57694eb46a0dc2427b8338ea05d4c98c98881d9",
	"special_tokens_map.json": "20e1ead0b0986b55989aa8116bafe63efcbae0daf8ed047970cd28b36aada4d9",
	"tokenizer.json": "58548a346eb073e5132bf7d8ad17dc6971bca36ade378ca4d2bfbc49bf60da2a",
	"tokenizer_config.json": "e2950960d2a3040a1938e6357c8c3628d531133c464286a7869b2e0504fcc59e"
	},
	"weight_bytes": 9873101418,
	"validation_status": "pending"
	}