Instructions to use DreamFoundries/gemma-4-E4B-it-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DreamFoundries/gemma-4-E4B-it-4bit 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("DreamFoundries/gemma-4-E4B-it-4bit")

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 DreamFoundries/gemma-4-E4B-it-4bit with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "DreamFoundries/gemma-4-E4B-it-4bit"

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": "DreamFoundries/gemma-4-E4B-it-4bit"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use DreamFoundries/gemma-4-E4B-it-4bit with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "DreamFoundries/gemma-4-E4B-it-4bit"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "DreamFoundries/gemma-4-E4B-it-4bit"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "DreamFoundries/gemma-4-E4B-it-4bit",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use DreamFoundries/gemma-4-E4B-it-4bit 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 "DreamFoundries/gemma-4-E4B-it-4bit"

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 DreamFoundries/gemma-4-E4B-it-4bit

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use DreamFoundries/gemma-4-E4B-it-4bit with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "DreamFoundries/gemma-4-E4B-it-4bit"

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 "DreamFoundries/gemma-4-E4B-it-4bit" \
  --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"

DreamFoundries commited on Jun 29

Commit

17c4646

verified ·

1 Parent(s): c220d48

Add files using upload-large-folder tool

Browse files

Files changed (3) hide show

README.md +4 -2
model.safetensors +2 -2
model.safetensors.index.json +1 -3

README.md CHANGED Viewed

@@ -24,11 +24,13 @@ This repository contains an MLX-LM conversion of [google/gemma-4-E4B-it](https:/
 - Quantization: MLX-LM affine quantization
 - Bits: 4-bit
 - Group size: 64
-- Local MLX folder size at upload time: 3.94 GiB
-- Local safetensors weight size at upload time: 3.91 GiB
 This Gemma conversion follows the MLX-LM Gemma 4 shared-KV topology and uses non-strict checkpoint loading so extra HF tensors outside that topology are discarded during conversion.
 ## Usage
 ```bash

 - Quantization: MLX-LM affine quantization
 - Bits: 4-bit
 - Group size: 64
+- Local MLX folder size at upload time: 3.98 GiB
+- Local safetensors weight size at upload time: 3.95 GiB
 This Gemma conversion follows the MLX-LM Gemma 4 shared-KV topology and uses non-strict checkpoint loading so extra HF tensors outside that topology are discarded during conversion.
+For mlx-swift compatibility, `per_layer_model_projection` was left unquantized while the rest of the eligible linear layers were quantized.
 ## Usage
 ```bash

model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e105100225815b521c6405c57ea23974530b868ddfea1e3ea9ec7ce1cf9fae38
-size 4198925877

 version https://git-lfs.github.com/spec/v1
+oid sha256:4d3184c7feccefb3ac857d73fc7d8a68297f30ff445a2ec757e49762e582ef54
+size 4238492977

model.safetensors.index.json CHANGED Viewed

@@ -1,6 +1,6 @@
 {
     "metadata": {
-        "total_size": 4198750292,
         "total_parameters": 7463013376
     },
     "weight_map": {
@@ -1355,8 +1355,6 @@
         "language_model.model.layers.9.self_attn.v_proj.scales": "model.safetensors",
         "language_model.model.layers.9.self_attn.v_proj.weight": "model.safetensors",
         "language_model.model.norm.weight": "model.safetensors",
-        "language_model.model.per_layer_model_projection.biases": "model.safetensors",
-        "language_model.model.per_layer_model_projection.scales": "model.safetensors",
         "language_model.model.per_layer_model_projection.weight": "model.safetensors",
         "language_model.model.per_layer_projection_norm.weight": "model.safetensors"
     }

 {
     "metadata": {
+        "total_size": 4238317652,
         "total_parameters": 7463013376
     },
     "weight_map": {
         "language_model.model.layers.9.self_attn.v_proj.scales": "model.safetensors",
         "language_model.model.layers.9.self_attn.v_proj.weight": "model.safetensors",
         "language_model.model.norm.weight": "model.safetensors",
         "language_model.model.per_layer_model_projection.weight": "model.safetensors",
         "language_model.model.per_layer_projection_norm.weight": "model.safetensors"
     }