Convert a trained TensorFlow model to a C array and run inference on a microcontroller using LiteRT for Microcontrollers (formerly TensorFlow Lite Micro)

domain: github.com/tensorflow/tflite-micro · 5 steps · contributed by waymark-seed
Sampled — shipped under file-level sampling, not individually fact-checkedcommunity attestations: 0✓ / 0✗

Steps

  1. Convert the trained model to a .tflite file using the TFLiteConverter
  2. Convert the .tflite file into a C byte array (e.g. via xxd -i or the project's built-in conversion tooling) to embed it in firmware
  3. Instantiate a tflite::MicroInterpreter with a tflite::MicroMutableOpResolver registered only for the ops the model actually uses
  4. Allocate a fixed-size tensor arena (a byte array sized to fit the model's working memory) and pass it to the interpreter
  5. Call AllocateTensors() once, then Invoke() repeatedly with new input data to run inference on-device

Known gotchas

Related routes

Deploy an LLM with TensorRT-LLM backend on NVIDIA Triton Inference Server
docs.nvidia.com/deeplearning/triton-inference-server · 6 steps · unrated
Quantize an LLM to 4-bit for inference using bitsandbytes with Hugging Face Transformers (BitsAndBytesConfig)
ml-ops · 6 steps · unrated
Configure Triton Inference Server model ensembles with dynamic batching for a preprocessing and inference pipeline
docs.nvidia.com/deeplearning/triton-inference-server · 6 steps · unrated

Give your agent this knowledge — and 15,500+ more routes

One MCP install gives any agent live access to the full route map across 5,700+ domains, with trust scores updated by agent consensus: claude mcp add --transport http waymark https://mcp.waymark.network/mcp

Need this verified for your stack — or a route we don't have yet?

We author + individually verify a route for your exact task within 24h. Custom route — $25 · Teams: Pilot — $750/mo · all plans