https://tigris-ml.dev/Recent content on TiGrISHugo -- gohugo.ioenCopyright (c) 2025-2026 raws labsThu, 11 Jun 2026 00:00:00 +0000https://tigris-ml.dev/blog/static-memory-planning/Thu, 11 Jun 2026 00:00:00 +0000https://tigris-ml.dev/blog/static-memory-planning/Peak working memory depends on the execution schedule and allocation, so the schedule can be optimized ahead of time and executed with bounded runtime arenas.https://tigris-ml.dev/blog/introducing-tigris/Thu, 23 Apr 2026 00:00:00 +0000https://tigris-ml.dev/blog/introducing-tigris/The model of your choice doesn’t fit in SRAM. You consider reducing its size or using a different model. TiGrIS tiles the computation instead and makes the exact model run on your target hardware.https://tigris-ml.dev/docs/getting-started/quickstart/Fri, 27 Feb 2026 00:00:00 +0000https://tigris-ml.dev/docs/getting-started/quickstart/End-to-end walkthrough: take an ONNX model that does not fit in your target’s SRAM, tile it with TiGrIS, and deploy it.https://tigris-ml.dev/docs/getting-started/installation/Mon, 01 Jan 0001 00:00:00 +0000https://tigris-ml.dev/docs/getting-started/installation/Python toolchain Requires Python 3.10 or later. pip install tigris-ml This pulls in all runtime dependencies automatically. Development install Clone the repo and install with dev extras for testing:https://tigris-ml.dev/docs/getting-started/core-concepts/Mon, 01 Jan 0001 00:00:00 +0000https://tigris-ml.dev/docs/getting-started/core-concepts/The key ideas behind how TiGrIS compiles and executes models on memory-constrained devices. Execution plan (.tgrs) A .tgrs file is a binary artifact containing the compiler-chosen operator schedule, stage and tiling strategy, tensor metadata, memory budget, and weights.https://tigris-ml.dev/docs/toolchain/tigris-analyze/Mon, 01 Jan 0001 00:00:00 +0000https://tigris-ml.dev/docs/toolchain/tigris-analyze/Check if an ONNX model fits your target’s memory constraints before compiling. Usage tigris analyze MODEL [OPTIONS] Options Flag Type Required Description MODEL path yes ONNX model file (.https://tigris-ml.dev/docs/toolchain/tigris-compile/Mon, 01 Jan 0001 00:00:00 +0000https://tigris-ml.dev/docs/toolchain/tigris-compile/Compile an ONNX model into a binary .tgrs execution plan for deployment on embedded devices. Usage tigris compile MODEL [OPTIONS] Options Flag Type Required Description MODEL path yes ONNX model file (.https://tigris-ml.dev/docs/toolchain/tigris-codegen/Mon, 01 Jan 0001 00:00:00 +0000https://tigris-ml.dev/docs/toolchain/tigris-codegen/Generate backend-specific C source code from a compiled .tgrs plan. The generated harness sets up runtime memory, selects the requested kernel dispatch, and loads the plan from the target’s normal deployment location.https://tigris-ml.dev/docs/toolchain/tigris-simulate/Fri, 27 Feb 2026 00:00:00 +0000https://tigris-ml.dev/docs/toolchain/tigris-simulate/Print a step-by-step execution trace for an ONNX model on the host, without target hardware. Use this to validate operator ordering, tensor lifetimes, and memory usage before flashing a device.https://tigris-ml.dev/docs/architecture/compilation-pipeline/Mon, 01 Jan 0001 00:00:00 +0000https://tigris-ml.dev/docs/architecture/compilation-pipeline/TiGrIS compiles an ONNX model into a .tgrs execution plan through seven stages. Each stage transforms the model representation, culminating in a binary that the C99 runtime executes directly on the target device.