AI study assistant for CUET, Banking, and Commerce students, powered by the Tiny AI model
A tiny byte-level multi-head content classifier (~1.5M params, ~200KB ONNX, <6ms). Classifies code, text, markup, config, images, binary, secrets, 62 code languages, 30 text languages, 90 MIME types from raw bytes — no tokenizer needed.
Lekhani: a 0.87 MB Hindi transliterator (type Hindi in Latin letters, get Devanagari) that runs in the browser or Node, on your own device. By Udgam Labs.
An 184M-parameter monolingual decision engine delivering calibrated distributions from one 22ms forward pass while matching models 2.3x its size.
Fracast-0 is an 85K-parameter pretrained Time Series Foundation Model, small enough to load directly in a web page.
Developed and pre-trained a 20.39M-parameter Punjabi GPT-style base model from scratch, including corpus preparation, tokenizer training, benchmark evaluation, and text generation, using a cleaned Punjabi corpus and local Apple Silicon GPU acceleration.
手搓零训练「随机数」GGUF 模型：不管输入什么，只吐一个均匀随机数字然后自己停。0/1 · 0-9 · 0-100 三个版本 × F16/Q8_0/Q4_K_M/Q4_K_S/Q3_K_M/Q2_K 六档量化，18 个文件全部实测通过 χ² 均匀性检验。
一个可以读完的微型 CNN 案例：规则老师采集 17 万帧 → 1.66M 参数小网络行为克隆 → CPU 1.3ms 实时玩 Flappy Bird，五档难度全通、地狱单局 141 分。PyTorch + Pygame，全流程可复现。
Bio-inspired CNN modeling M/P/K pathways of the Lateral Geniculate Nucleus. From 76KB/15.5K params (33 FPS on Raspberry Pi) to full-scale and temporal variants. Matches ResNet18 with 52x fewer parameters. No pretraining required.
A tiny scorer that tells a big AI what it forgot to do. CPU-only, 1 core, ~1GB RAM. Not a language model.
AI document generator powered by the Tiny AI model. Creates Word, PDF, PPTX, XLSX, and Markdown files completely client-side with real-time response streaming.
Small Model Evaluation
To associate your repository with the tiny-model topic, visit your repo's landing page and select "manage topics."