Description
LLM-8850 Kit is a high-performance AI accelerator kit designed for edge AI and embedded computing scenarios. It consists of the LLM-8850 Card AI accelerator and the LLM-8850 PiHat adapter board. The core LLM-8850 Card is based on the Axera AX8850 SoC and adopts the M.2 M-Key 2242 form factor, delivering up to 24 TOPS @ INT8 of computing power in a compact size. It efficiently accelerates multimodal large-model inference and video analytics workloads. The card integrates an active cooling system with a micro blower fan + CNC aluminum fin heatsink, with intelligent speed control driven by the onboard EC according to a temperature–current curve. This ensures stable low-temperature operation even under long-term full load or in enclosed chassis environments, preventing thermal throttling. The accompanying LLM-8850 PiHat adapter board provides a 2.54-40P bus and a USB Type-C PD power input, and connects to the Raspberry Pi 5 via a PCIe expansion FPC adapter, enabling seamless integration of high-performance AI acceleration on the Raspberry Pi platform for building compact and efficient edge intelligence systems.
Features
- Ultra-compact form factor: M.2 M-Key 2242 size
- High-performance NPU: 24 TOPS @ INT8, octa-core Cortex-A55 1.7 GHz CPU
- Intelligent cooling / power: onboard blower fan + CNC aluminum integrated heatsink, EC-monitored temperature–current–speed closed loop
- 8GB (4+4) high-bandwidth memory: 64-bit LPDDR4x, 4266 Mbps
- Hardware video engine: 8K @ 30 fps H.264/H.265 encoding, 8K @ 30 fps decoding, supports 16-channel 1080P parallel decoding
- Secure boot & encryption: AES / DES / 3DES / SHA-256 hardware security module
- Native AXCL support: one-click full-stack model deployment for CNN, Transformer, CLIP, Whisper, Llama3.2, Qwen3, InternVL3, etc., with H.264/H.265 simultaneous encode/decode transcoding
- USB Type-C PD power input:
- USB PD3.0 100W protocol
- DC 24V overvoltage protection
- Power capability requirement > 9V@3A (27W)
Includes
LLM-8850 Kit (8G) (SKU:AI-002-8G)
- 1 x LLM-8850 Card (8G)
- 1 x LLM-8850 PiHat
- 1 x 2.54-40P Female Header
- 1 x Screw and Standoff Accessory Pack
Applications
- Edge AI inference deployment
- Embedded video analytics
- Raspberry Pi AI computing expansion
- Multimodal large-model implementation
Specifications
| Specification | Parameter |
|---|---|
| SoC | Axera AX8850 |
| CPU | Octa-core Cortex-A55 1.7 GHz |
| NPU | 24 TOPS @ INT8 |
| Video Encoder | 8K @ 30 fps H.264/H.265 encoding, supports scaling / cropping |
| Video Decoder | 8K @ 30 fps H.264/H.265 decoding, supports 16-channel 1080P parallel decoding, scaling / cropping |
| Memory | Optional: 8GB 64-bit LPDDR4x, 4266 Mbps (4+4) / 4GB 64-bit LPDDR4x, 4266 Mbps (2+2) |
| Storage | 32Mbits QSPI NOR Flash (bootloader only) |
| Form Factor | M.2 M-Key 2242, PCIe 2.0 x2 |
| Cooling | Micro blower fan + CNC aluminum integrated heatsink, EC intelligent thermal control |
| Operating Temperature | 0 ~ 60 °C |
| Full-load Temperature at Room Temp | 70 °C |
| Adapter Board PD Power | USB PD3.0 100W protocol (DC 9V / 12V / 20V) |
| Adapter Board Load Capability | Output to Raspberry Pi 5: DC 5V@4A; Output to M.2 accelerator card: DC 3.3V@6A |
| LLM-8850 Card Power | 7W @ 3.3V |
| Power Requirement | > 9V@3A (27W) |
| Product Size | LLM-8850 Card: 42.6 x 24.0 x 9.7mm LLM-8850 PiHat: 65.0 x 58.0 x 12.7mm |
| Product Weight | 31.9g |
| Package Size | 165.0 x 90.0 x 17.5mm |
| Gross Weight | 85.0g |