[ICLR2022] Efficient Split-Mix federated learning for in-situ model customization during both training and testing time
-
Updated
Apr 12, 2023 - Python
#
tiny-ml
Here are 14 public repositories matching this topic...
Benchmarking machine learning inferencing on embedded hardware.
raspberry-pi benchmark machine-learning benchmarking-framework movidius jetson-nano coral-tpu tiny-ml small-ai
-
Updated
Jul 18, 2025 - Python
Source code of “Agile Reinforcement Learning for Real-Time Task Scheduling in Edge Computing” (CAIAC 2025)
reinforcement-learning embedded-systems embedded-devices task-scheduler real-time-systems scheduling-algorithms edge-computing task-scheduling tiny-ml real-time-applications action-masking informed-exploration
-
Updated
Mar 12, 2026 - Python
An editable, auditable 807K-param byte-level LLM: CRUD single facts with provable per-edit locality, and abstain when unsure instead of guessing. CPU, offline.
machine-learning crud cpu pytorch grace knowledge-base language-model byte-level interpretability rome serac abstention tiny-ml selective-prediction llm model-editing knowledge-editing memit
-
Updated
Jun 20, 2026 - Python
PXG-Tiny Ai - pixel sprites game asset generator. A 483K-param open-weights model that turns plain English prompts into 16×16 pixel-art sprites — ~1 MB NumPy weights, CPU-only, fully offline.
gamedev machine-learning deep-learning sprites pixel-art numpy game-development roguelike pytorch image-generation text-to-image game-assets awesome-ai model-weights asset-generator ai-art tiny-ml generative-ai text-2-image open-weights
-
Updated
Aug 27, 2026 - Python
Tiny CNN cat/dog classifier for RISC-V edge AI, with PyTorch training, quantized firmware export, and Renode simulation benchmarks.
fpga embedded-systems computer-engineering academic-project fixed-point quantization hardware-acceleration risc-v cifar-10 cnn-classification renode edge-ai tiny-ml
-
Updated
May 6, 2026 - Python
A new communication paradigm proposal for restricted bandwidth and fragile channel conditions in NTN.
-
Updated
Apr 24, 2026 - Python
Ingenuity is an optimized inference engine and benchmarking tool for TinyML models on embedded IoT devices.
-
Updated
Nov 19, 2025 - Python
Generalized Kuramoto coupling topology for language semantics — distill, converge, decode
machine-learning deep-learning neural-network tensorflow pytorch artificial-intelligence knowledge-distillation model-compression efficient-inference phase-synchronization edge-ai on-device-ml kuramoto tiny-ml sparse-inference coupled-oscillators llm-inference transformer-alternative semantic-topology distillation-without-training
-
Updated
Sep 13, 2026 - Python
Sub-3K parameter MNIST benchmark - 98%+ accuracy with tiny MLP + wavelet scattering transform. Tiny machine learning research.
c benchmark machine-learning deep-learning neural-network pytorch mnist knowledge-distillation edge-ai tinyml tiny-ml
-
Updated
Aug 17, 2026 - Python
-
Updated
Oct 4, 2026 - Python
TRM-inspired tiny recursive reasoning network — 44k params solving Sudoku on CPU with deep supervision, honest baseline comparison (PyTorch + FastAPI)
-
Updated
Sep 20, 2026 - Python
Tiny open-source model (1.2 MB) that understands percentage questions in English, Portuguese and Spanish and returns the right calculation. Runs in the browser in ~1 ms.
javascript python calculator machine-learning natural-language-processing browser percentage nlu transformer spanish portuguese on-device percentage-calculator tiny-ml porcentaje porcentagem tool-calling
-
Updated
Oct 2, 2026 - Python
Goal: a language model that fits on a real 3.5" floppy. 1 474 560 bytes is empty FAT12 geometry, not a packed model. Sub-bit core codes vs scalar quantization; quality not claimed.
-
Updated
Oct 4, 2026 - Python
Add this topic to your repo
To associate your repository with the tiny-ml topic, visit your repo's landing page and select "manage topics."