Getting Started with SL2600
Set up your board and run your first edge AI examples.
Image Classification
Recognize and label images with a classifier on the SL2600 NPU.
Object Detection
Find and locate objects in images with YOLO on the SL2600 NPU.
Bring Your Own Model
Convert your own model with the Torq compiler and run it on the SL2600 NPU.
Getting Started with SL1600
Set up your board and run your first edge AI examples.
Image Classification
Recognize and label images with a classifier on the SL1600 NPU.
Object Detection
Find and locate objects in images with YOLO on the SL1600 NPU.
Speech-To-Text
Transcribe speech on-device with Whisper and Moonshine.
Text-to-Speech
Turn written text into natural-sounding speech on-device with Piper.
Embeddings
Learn how text embeddings work and generate them on-device with Synaptics Astra.
AI Text Assistant
Build a Q&A assistant that picks answers using on-device embeddings.
AI Tool Calling
Let an on-device AI model control an LED and system functions via tool calling.
Large Language Models
Run small large language models locally on Astra.
Bring Your Own Model
Compile your own model with the SyNAP toolkit and run it on the SL1600 NPU.
NPU Acceleration from C++
Build C++ apps that run models on the NPU with the SyNAP framework.
Modifying SyNAP Framework
Modify the SyNAP framework to build your own C++ inference apps.
Build llama-cpp Python wheel
Build llama-cpp-python on the board for local LLM inference from Python.
LLM on Astra
Run TinyLlama natively on Astra Machina with llama.cpp.
Cross Compile LLM
Cross-compile llama.cpp on a host machine for Astra Machina.
VLM on Astra
Run vision language models that understand images and text with llama.cpp.
Llamafile on Astra
Run Llama 3.2 1B on Astra as a single llamafile executable.
Whisper Speech-to-Text
Transcribe a WAV file on-device with whisper.cpp.
Whisper Streaming
Transcribe live microphone audio in real time with whisper.cpp.
Benchmark YOLO models
Measure the inference speed of pre-trained YOLO models on Astra.
Roboflow custom YOLO11
Fine-tune YOLO11 for PPE detection with Roboflow and deploy it to Astra.
Ultralytics YOLO for edge
Export an Ultralytics YOLOv8 pose model and compile it for the Astra NPU.
Evaluate Model Compatibility
Check your model's compatibility with the SR100 NPU, no hardware needed.
Evaluate Vision Models
Evaluate pre-optimized vision models for SR-Series MCUs in a Hugging Face Space.
Torq Introduction
An overview of the Torq Edge AI platform for running neural networks on the SL2610 NPU.
SyNAP Introduction
What the SyNAP toolkit does and how it runs models on the Astra NPU and GPU.
SyNAP Inference
Run the precompiled sample models and apps that ship on the board.
SyNAP Model Compilation
Compile TFLite, ONNX and other models into NPU-optimized SyNAP models.
SyNAP Conversion Metafile
Customize layout, quantization, preprocessing and delegates with a YAML metafile.
SyNAP Model Quantization
Quantize models to 8 or 16 bits so they run efficiently on the NPU.
SyNAP Framework API
Run precompiled networks from C++ with the SyNAP Network class.
SyNAP Android Applications
Use SyNAP from Android apps through JNI for zero-copy NPU inference.
SyNAP JIT Compilation
Run TensorFlow Lite models directly with just-in-time compilation.
SyNAP NPU Operators
Neural network operators, tensor types and engines the NPU supports.
SyNAP Statistics and Usage
Monitor NPU usage and inference statistics through sysfs.
