Add the M5Stack LLM-8850 M.2 Axera AX8850 AI accelerator adapter

[Hedda]

M5Stack LLM-8850 M.2 Axera AX8850 AI accelerator adapter with both AI object detection + video transcoding (video decoder/encoder):

• LLM-8850 M.2 Axera AX8850 AI accelerator adapter for $99 from M5Stack compatibility with Frigate NVR for AI object detection + video transcoding? #770

  • https://www.cnx-software.com/2025/10/03/m5stack-llm-8850-card-an-m-2-m-key-ai-accelerator-module-based-on-axera-ax8850-24-tops-soc/

    • https://docs.m5stack.com/en/ai_hardware/LLM-8850_Card

      • https://docs.m5stack.com/en/guide/ai_accelerator/llm-8850/m5_llm_8850_software_install

[Hedda]

FYI, also see "Radxa AICore AX-M1" M.2 2280 M Key form-factor AI Acceleration Module (M72T) based on same Axera AX8850 SoC:

• Add the Radxa AICore AX-M1" M.2 2280 AI Acceleration Module based Axera AX8850 SoC #774

  • https://radxa.com/products/aicore/ax-m1/

    • https://radxa.com/products/aicore/

Note! While not 100% sure I believe Axera AX8850 is the same as the Axera M76H SoC on their site targeting "smart driving"(?):

• https://en.axera-tech.com/Product/108.html

  • https://en.axera-tech.com/filespath/files/20240914105017.pdf

RapidAnalysis posted a demo-video of a preview unit from Radxa that looks like a final protoype that has "radxa AICore M72T V1.0" printed on it. Video show simple demo of DeepSeek-R1-Qwen-7B and SmolLM2-360M-Instruct large language model(s):

• https://www.youtube.com/watch?v=4dGTC-YSq1g

Radxa has a support listing for Large Language Models, small and large Vision models, Speech Models, Text-to-image Generation models:

Large Language Models:

• DeepSeek-R1-Distill
• Qwen2.5
• Qwen3
• MiniCPM4
• SmolLM3
• Llama3.2
• Gemma2
• Phi3

Vision Large Models:

• InternVL3
• Qwen2.5-VL
• SmolVLM2
• CLIP
• YOLOWorldv2

Speech Models:

• Whisper
• SenseVoice
• MeloTTS

Text-to-image Generation Model:

• Stable Diffusion v1.5

Vision Model:

• YOLO Family
• Depth-Anything-V2
• Real-ESRGAN
• MixFormerV2
• LivePortrait