M5Stack LLM (large language model) Module Kit (AX630C)
📦 Location of the warehouse: Munich
🚚 We will ship your order within 24 hours after the order is taken.
- Pre-Ordering: If an item is unavailable, you can pre-order it. We'll dispatch once stock is replenished, usually within 10 working days.
- Express Shipping: Need it faster? Contact us via email or through our live chat support, and we'll arrange expedited delivery for you.
Free Shipping from 30€ in Germany, 50€ in the EU!
- EU Countries Delivery Time: 4~8 working days.
- Other Countries Delivery Time: 2~2.5 weeks.
Free shipping for you
Free Shipping from 30€ in Germany, 50€ in the EU!
Trustpilot rating: 4.4/5.0
Thank you for your support - rate us on Trustpilot!

M5Stack LLM (large language model) Module Kit (AX630C)
Module LLM Kit is a smart modular kit focused on offline AI inference and data communication interface applications. It integrates the Module LLM and Module13.2 LLM Mate modules to meet the offline AI inference and data interaction requirements across various scenarios.
Module LLM is an integrated offline large language model (LLM) inference module designed specifically for terminal devices that require efficient and intelligent interaction. Whether for smart home applications, voice assistants, or industrial control, Module LLM delivers a smooth and natural AI experience without relying on the cloud, ensuring privacy, security, and stability.
Module13.2 LLM Mate Module provides a variety of interface functions to facilitate system integration and expansion. It achieves stacked power supply with Module LLM via the M5BUS interface; its built-in CH340N USB conversion chip offers USB-to-serial debugging functionality, while the Type-C interface is used for USB log output. Additionally, the RJ45 interface works with the onboard network transformer to extend to a 100 Mbps Ethernet port and core serial port (supporting SBC applications); the FPC-8P interface connects directly to Module LLM, ensuring stable serial communication; furthermore, an HT3.96*9P solder pad is reserved for DIY expansion.
The Module LLM module integrates the StackFlow framework along with the Arduino/UiFlow libraries, allowing edge intelligence to be implemented with just a few lines of code. Powered by the advanced AiXin AX630C SoC processor and featuring a high-efficiency NPU delivering 3.2 TOPS with native support for Transformer models, it effortlessly handles complex AI tasks. Equipped with 4GB LPDDR4 memory (1GB for user applications and 3GB dedicated to hardware acceleration) and 32GB eMMC storage, it supports parallel multi-model loading and chained inference, ensuring smooth multitasking. With an operating power consumption of only about 1.5W, it is far more energy efficient than similar products, making it ideal for long-term operation.
Module LLM is compatible with multiple models and comes pre-installed with the Qwen2.5-0.5B large language model, featuring built-in functions including KWS (wake word), ASR (speech recognition), LLM (large language model), and TTS (text-to-speech). It also supports apt-based rapid updates of software and model packages. By installing the openai-api plugin, it becomes compatible with the OpenAI standard API, supporting chat, conversation completion, speech-to-text, and text-to-speech among various application modes. The official apt repository offers abundant large model resources—including deepseek-r1-distill-qwen-1.5b, InternVL2_5-1B-MPO, Llama-3.2-1B, Qwen2.5-0.5B, and Qwen2.5-1.5B—as well as text-to-speech models (whisper-tiny, whisper-base, melotts) and visual models (such as yolo11 and other SOTA models). The repository is continuously updated to support the most cutting-edge model applications, meeting the demands of complex AI tasks.
Module LLM Kit is plug-and-play, and when paired with the M5 host, it provides an instant AI interactive experience. Users can seamlessly integrate it into existing smart devices without cumbersome setup, quickly enabling intelligent features and enhancing device performance. This product is ideal for offline voice assistants, text-to-speech conversion, smart home control, interactive robots, and more.
Features
- Offline inference, 3.2 TOPS at INT8 precision
- Integrated KWS (wake word), ASR (speech recognition), LLM (large language model), TTS (text-to-speech)
- Parallel multi-model processing
- Onboard 32GB eMMC storage and 4GB LPDDR4 memory
- Onboard microphone and speaker
- Serial communication
- SD card firmware upgrade
- Supports ADB debugging
- RGB status LED
- Built-in Ubuntu system
- Supports OTG functionality
Includes
- 1 x Module LLM
- 1 x Module LLM Mate
- 2 x FPC-8P Wire
Applications
- Offline voice assistants
- Text-to-speech conversion
- Smart home control
- Interactive robots
Specifications
Specification | Parameter |
---|---|
Processor SoC | AX630C@Dual Cortex A53 1.2 GHz MAX.12.8 TOPS @INT4, 3.2 TOPS @INT8 |
Memory | 4GB LPDDR4 (1GB system memory + 3GB dedicated to hardware acceleration) |
Storage | 32GB eMMC5.1 |
Communication | Serial communication, default baud rate 115200@8N1 (adjustable) |
Microphone | MSM421A |
Audio Driver | AW8737 |
Speaker | 8Ω@1W, size: 2014 cavity speaker |
Built-in Functions | KWS (wake word), ASR (speech recognition), LLM (large language model), TTS (text-to-speech) |
RGB LED | 3x RGB LED@2020, driven by LP5562 (status indicator) |
Power Consumption | No load: 5V@0.5W, Full load: 5V@1.5W |
Button | Used to enter firmware download mode |
Upgrade Interface | SD card/Type-C port |
Conversion Chip | CH340N |
Ethernet Interface | RJ45 interface with onboard network transformer (11FB-05NL SOP-16) |
Serial Interfaces | FPC-8P interface, Type-C interface, RJ45 interface |
DIY Expansion | HT3.96*9P solder pad |
Operating Temp. | 0-40°C |
Product Size | Module LLM: 54.0 x 54.0 x 13.0mm Module13.2 LLM Mate: 54 x 54 x 19.7mm |
Package Size | Module LLM: 192.0 x 95.0 x 17.0mm Module13.2 LLM Mate: 192.0 x 95.0 x 21.0mm |
Product Weight | Module LLM: 17.4g Module13.2 LLM Mate: 19.2g |
Package Weight | Module LLM: 32.0g Module13.2 LLM Mate: 34.8g |
Effective Date: March 27, 2025
At OpenELAB, we are committed to providing accurate product descriptions and delivering high-quality products. If you experience any issues with your purchase, please notify us within 60 days of the shipment date.
1. Warranty Disclaimer
Unless explicitly stated in a specific agreement or purchase order, OpenELAB makes no warranties—express or implied—regarding the fitness, suitability, or merchantability of any product for a particular purpose or special use.
2. Warranty Policy
Warranty Period
The standard warranty period for products sold by OpenELAB is 12 months from the date of delivery.
Under this warranty, OpenELAB will make a reasonable effort to repair or replace products free of charge if defects are due to issues with raw materials, design, or manufacturing.
If the defect qualifies under this policy but falls outside the covered scope, reasonable charges may apply for repair or replacement.
After the warranty period, support may still be available at a reasonable fee.
Warranty Exclusions
The warranty does not apply in the following cases:
- The product is beyond the 12-month warranty period.
- Damage due to misuse, improper maintenance or storage, or failure to follow the user guide or manual.
- Issues caused by Force Majeure events (e.g., natural disasters, war, legal changes, etc.).
- Unauthorized repair attempts by the customer.
- The product is not defective but has been opened and is non-returnable by policy.
3. Returns & Refunds
If you're not satisfied with your purchase, you may be eligible for a return or exchange:
- Return for Refund: Within 15 days of receipt (unopened).
- Return for Replacement: Within 30 days of receipt (damaged or defective).
- Exchange Window: Up to 30 days from receipt.
Please email us at info@openelab.io for a return authorization before sending anything back.
Return Conditions
- Items must be in original, clean, and mint condition with packaging intact.
- Returns of damaged, used, scratched, modified, or programmed items will not be accepted.
- Include the receipt, a brief note with your name, order number, and reason for return.
Shipping & Fees
- Customers are responsible for return shipping to our warehouses in Europe, the US, or China.
- Refunds may be subject to additional shipping fees unless the return is due to our error.
- A flat €8 shipping fee will be deducted for returned items that originally qualified for free shipping.
Payment Platform Fees
For refunds processed through third-party platforms such as PayPal or Braintree, a standard 3.5% service fee may be deducted to cover payment processing costs.