Instructions to use mlx-community/gemma-4-e4b-it-8bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use mlx-community/gemma-4-e4b-it-8bit with MLX:
# Make sure mlx-vlm is installed # pip install --upgrade mlx-vlm from mlx_vlm import load, generate from mlx_vlm.prompt_utils import apply_chat_template from mlx_vlm.utils import load_config # Load the model model, processor = load("mlx-community/gemma-4-e4b-it-8bit") config = load_config("mlx-community/gemma-4-e4b-it-8bit") # Prepare input image = ["http://images.cocodataset.org/val2017/000000039769.jpg"] prompt = "Describe this image." # Apply chat template formatted_prompt = apply_chat_template( processor, config, prompt, num_images=1 ) # Generate output output = generate(model, processor, formatted_prompt, image) print(output) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Pi
How to use mlx-community/gemma-4-e4b-it-8bit with Pi:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "mlx-community/gemma-4-e4b-it-8bit"
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": "mlx-community/gemma-4-e4b-it-8bit" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Hermes Agent
How to use mlx-community/gemma-4-e4b-it-8bit 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 "mlx-community/gemma-4-e4b-it-8bit"
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 mlx-community/gemma-4-e4b-it-8bit
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use mlx-community/gemma-4-e4b-it-8bit with OpenClaw:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "mlx-community/gemma-4-e4b-it-8bit"
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 "mlx-community/gemma-4-e4b-it-8bit" \ --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"
Fix chat template: emit multimodal placeholders in tool messages
Browse filesMirrors the upstream google/gemma-4-*-it commit fcf2302 - adds emission of
<|image|>, <|audio|>, and <|video|> placeholders for tool-message content
parts so multimodal tool responses are not silently dropped.
- chat_template.jinja +25 -9
chat_template.jinja
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
-
{%- macro format_parameters(properties, required) -%}
|
| 2 |
{%- set standard_keys = ['description', 'type', 'properties', 'required', 'nullable'] -%}
|
| 3 |
{%- set ns = namespace(found_first=false) -%}
|
| 4 |
{%- for key, value in properties | dictsort -%}
|
| 5 |
{%- set add_comma = false -%}
|
| 6 |
-
{%- if key not in standard_keys -%}
|
| 7 |
{%- if ns.found_first %},{% endif -%}
|
| 8 |
{%- set ns.found_first = true -%}
|
| 9 |
{{ key }}:{
|
|
@@ -65,7 +65,7 @@
|
|
| 65 |
{%- elif value is mapping -%}
|
| 66 |
{%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}
|
| 67 |
properties:{
|
| 68 |
-
{{- format_parameters(value, value['required'] | default([])) -}}
|
| 69 |
}
|
| 70 |
{%- endif -%}
|
| 71 |
{%- if value['required'] -%}
|
|
@@ -178,18 +178,21 @@
|
|
| 178 |
{#- Handle System/Tool Definitions Block -#}
|
| 179 |
{%- if (enable_thinking is defined and enable_thinking) or tools or messages[0]['role'] in ['system', 'developer'] -%}
|
| 180 |
{{- '<|turn>system\n' -}}
|
| 181 |
-
|
| 182 |
{#- Inject Thinking token at the very top of the FIRST system turn -#}
|
| 183 |
{%- if enable_thinking is defined and enable_thinking -%}
|
| 184 |
{{- '<|think|>\n' -}}
|
| 185 |
{%- set ns.prev_message_type = 'think' -%}
|
| 186 |
{%- endif -%}
|
| 187 |
-
|
| 188 |
{%- if messages[0]['role'] in ['system', 'developer'] -%}
|
| 189 |
-
{
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 190 |
{%- set loop_messages = messages[1:] -%}
|
| 191 |
{%- endif -%}
|
| 192 |
-
|
| 193 |
{%- if tools -%}
|
| 194 |
{%- for tool in tools %}
|
| 195 |
{{- '<|tool>' -}}
|
|
@@ -198,7 +201,6 @@
|
|
| 198 |
{%- endfor %}
|
| 199 |
{%- set ns.prev_message_type = 'tool' -%}
|
| 200 |
{%- endif -%}
|
| 201 |
-
|
| 202 |
{{- '<turn|>\n' -}}
|
| 203 |
{%- endif %}
|
| 204 |
|
|
@@ -293,6 +295,15 @@
|
|
| 293 |
{%- endif -%}
|
| 294 |
{%- endfor -%}
|
| 295 |
{{- format_tool_response_block(ns_tname.name, ns_txt.s) -}}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 296 |
{%- else -%}
|
| 297 |
{{- format_tool_response_block(ns_tname.name, tool_body) -}}
|
| 298 |
{%- endif -%}
|
|
@@ -302,6 +313,7 @@
|
|
| 302 |
{%- endfor -%}
|
| 303 |
{%- endif -%}
|
| 304 |
|
|
|
|
| 305 |
{%- if message['content'] is string -%}
|
| 306 |
{%- if role == 'model' -%}
|
| 307 |
{{- strip_thinking(message['content']) -}}
|
|
@@ -328,10 +340,14 @@
|
|
| 328 |
{%- endif -%}
|
| 329 |
{%- endfor -%}
|
| 330 |
{%- endif -%}
|
|
|
|
|
|
|
|
|
|
|
|
|
| 331 |
|
| 332 |
{%- if ns.prev_message_type == 'tool_call' and not ns_tr_out.flag -%}
|
| 333 |
{{- '<|tool_response>' -}}
|
| 334 |
-
{%- elif not (ns_tr_out.flag and not
|
| 335 |
{{- '<turn|>\n' -}}
|
| 336 |
{%- endif -%}
|
| 337 |
{%- endif -%}
|
|
|
|
| 1 |
+
{%- macro format_parameters(properties, required, filter_keys=false) -%}
|
| 2 |
{%- set standard_keys = ['description', 'type', 'properties', 'required', 'nullable'] -%}
|
| 3 |
{%- set ns = namespace(found_first=false) -%}
|
| 4 |
{%- for key, value in properties | dictsort -%}
|
| 5 |
{%- set add_comma = false -%}
|
| 6 |
+
{%- if not filter_keys or key not in standard_keys -%}
|
| 7 |
{%- if ns.found_first %},{% endif -%}
|
| 8 |
{%- set ns.found_first = true -%}
|
| 9 |
{{ key }}:{
|
|
|
|
| 65 |
{%- elif value is mapping -%}
|
| 66 |
{%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}
|
| 67 |
properties:{
|
| 68 |
+
{{- format_parameters(value, value['required'] | default([]), filter_keys=true) -}}
|
| 69 |
}
|
| 70 |
{%- endif -%}
|
| 71 |
{%- if value['required'] -%}
|
|
|
|
| 178 |
{#- Handle System/Tool Definitions Block -#}
|
| 179 |
{%- if (enable_thinking is defined and enable_thinking) or tools or messages[0]['role'] in ['system', 'developer'] -%}
|
| 180 |
{{- '<|turn>system\n' -}}
|
|
|
|
| 181 |
{#- Inject Thinking token at the very top of the FIRST system turn -#}
|
| 182 |
{%- if enable_thinking is defined and enable_thinking -%}
|
| 183 |
{{- '<|think|>\n' -}}
|
| 184 |
{%- set ns.prev_message_type = 'think' -%}
|
| 185 |
{%- endif -%}
|
|
|
|
| 186 |
{%- if messages[0]['role'] in ['system', 'developer'] -%}
|
| 187 |
+
{%- if messages[0]['content'] is string -%}
|
| 188 |
+
{{- messages[0]['content'] | trim -}}
|
| 189 |
+
{%- elif messages[0]['content'] is sequence -%}
|
| 190 |
+
{%- for item in messages[0]['content'] -%}
|
| 191 |
+
{{- item['text'] | trim + ' '-}}
|
| 192 |
+
{%- endfor -%}
|
| 193 |
+
{%- endif -%}
|
| 194 |
{%- set loop_messages = messages[1:] -%}
|
| 195 |
{%- endif -%}
|
|
|
|
| 196 |
{%- if tools -%}
|
| 197 |
{%- for tool in tools %}
|
| 198 |
{{- '<|tool>' -}}
|
|
|
|
| 201 |
{%- endfor %}
|
| 202 |
{%- set ns.prev_message_type = 'tool' -%}
|
| 203 |
{%- endif -%}
|
|
|
|
| 204 |
{{- '<turn|>\n' -}}
|
| 205 |
{%- endif %}
|
| 206 |
|
|
|
|
| 295 |
{%- endif -%}
|
| 296 |
{%- endfor -%}
|
| 297 |
{{- format_tool_response_block(ns_tname.name, ns_txt.s) -}}
|
| 298 |
+
{%- for part in tool_body -%}
|
| 299 |
+
{%- if part.get('type') == 'image' -%}
|
| 300 |
+
{{- '<|image|>' -}}
|
| 301 |
+
{%- elif part.get('type') == 'audio' -%}
|
| 302 |
+
{{- '<|audio|>' -}}
|
| 303 |
+
{%- elif part.get('type') == 'video' -%}
|
| 304 |
+
{{- '<|video|>' -}}
|
| 305 |
+
{%- endif -%}
|
| 306 |
+
{%- endfor -%}
|
| 307 |
{%- else -%}
|
| 308 |
{{- format_tool_response_block(ns_tname.name, tool_body) -}}
|
| 309 |
{%- endif -%}
|
|
|
|
| 313 |
{%- endfor -%}
|
| 314 |
{%- endif -%}
|
| 315 |
|
| 316 |
+
{%- set captured_content -%}
|
| 317 |
{%- if message['content'] is string -%}
|
| 318 |
{%- if role == 'model' -%}
|
| 319 |
{{- strip_thinking(message['content']) -}}
|
|
|
|
| 340 |
{%- endif -%}
|
| 341 |
{%- endfor -%}
|
| 342 |
{%- endif -%}
|
| 343 |
+
{%- endset -%}
|
| 344 |
+
|
| 345 |
+
{{- captured_content -}}
|
| 346 |
+
{%- set has_content = captured_content | trim | length > 0 -%}
|
| 347 |
|
| 348 |
{%- if ns.prev_message_type == 'tool_call' and not ns_tr_out.flag -%}
|
| 349 |
{{- '<|tool_response>' -}}
|
| 350 |
+
{%- elif not (ns_tr_out.flag and not has_content) -%}
|
| 351 |
{{- '<turn|>\n' -}}
|
| 352 |
{%- endif -%}
|
| 353 |
{%- endif -%}
|