# ShinMegamiBoson/lookatmyquant > A post-training quantization compiler that allocates bits per row by measured output damage, producing GGUF/MLX files that keep lower KL than Q4_K_M at the same size. Ships an Apple Silicon runtime (lamq) for running the compressed models. - Magnitude: 7.7 out of 10 — Early signal - Stars: 461 total · +194 stars measured 2026-11-22, 00:09–14:25 UTC, ≈ 323 by evening - Star trust: star growth looks organic - Category: Language models · Language: Python · License: MIT · Created: 2026-10-31 · Last push: 2026-11-22 - GitHub: https://github.com/ShinMegamiBoson/lookatmyquant · Homepage: https://lookatmyquant.pages.dev · Page: https://gitnova.dev/en/r/ShinMegamiBoson/lookatmyquant ## Useful for - Compress a large model to a target file size with lower KL than Q4_K_M - Run a quantized model locally on Apple Silicon via the lamq CLI - Inspect a packed model's format and dimensions without loading weights ## How to get started ```sh pip install git+https://github.com/ShinMegamiBoson/lookatmyquant.git ``` Copied as is from the “install and use the apple silicon runtime” section of the README. GitNova has not run these commands: check them in the original before running: https://github.com/ShinMegamiBoson/lookatmyquant#readme ## Why it’s here - Star-counter measurements on 2026-11-22 (UTC), 00:09–14:25: 267 → 461 stars (+194). This is the change over that interval. - Estimated end-of-day forecast: about +323 stars, using observed gains and the previous day. - Before this, the repository barely got any stars — about 1 per day. - The repository is 22 days old and already has 461 stars. - Hacker News: “Inkling-2-Mini in 5.4 GB Memory” — 501 points, 20 h ago. - About 24 forks a day — people are taking the code. - Recent forks include notable developers: @mbrukman (404 followers). ## Star trust Star growth looks organic. Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer. ## Numbers - Forks: 19 - Issues and pull requests: 11 - Watchers: 1 - Average over the last week: 96 per day - Usual pace: 1 per day - Stars in the last hour (measured): 20 - Latest release: lm2-qwen35-4b-1.2gb (2026-11-12) ## Stars per day, last 23 days (oldest → newest, today is partial) 2026-10-31 … 2026-11-22: 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 4, 0, 0, 0, 0, 0, 0, 1, 2, 344, 194 ## Magnitude by day, last 2 days 11-21 6.5, 11-22 7.7 ## Hacker News - Inkling-2-Mini in 5.4 GB Memory — 501 points, 125 comments: https://news.ycombinator.com/item?id=50591853 ## Spotted in now - Spotted on Hacker News ## More in this category 1. **jingyaogong/minimind** — 6.9 · Breakout · Language models · Python · +302 stars measured 2026-11-22, 00:09–14:26 UTC, ≈ 491 by evening An open-source educational project for training a tiny 64M-parameter LLM (MiniMind) from scratch in PyTorch, covering the full pipeline: pretrain, SFT, LoRA, RLHF, RLAIF and Tool Use. Full card: https://gitnova.dev/en/r/jingyaogong/minimind.md 2. **sybil-solutions/inference-optimisation-skill** — 6.3 · Peaking · Language models · Python · +228 stars measured 2026-11-22, 00:09–14:26 UTC, ≈ 380 by evening Agent skill for hill-climbing the speed of a local LLM server: decode, speculative decoding, prefill, KV cache, startup and TTFT on any hardware and quantization. Full card: https://gitnova.dev/en/r/sybil-solutions/inference-optimisation-skill.md 3. **openvinotoolkit/openvino** — 5.4 · Breakout · Language models · C++ · +103 stars measured 2026-11-22, 00:09–14:26 UTC, ≈ 167 by evening OpenVINO is an open-source toolkit for optimizing and deploying deep learning model inference on Intel CPU, GPU and NPU. It supports models from PyTorch, TensorFlow, ONNX, Keras, PaddlePaddle, JAX/Flax and Hugging Face with C++, Python, C… Full card: https://gitnova.dev/en/r/openvinotoolkit/openvino.md 4. **sous-labs/songe** — 5.4 · Early signal · Language models · Python · +40 stars measured 2026-11-22, 00:10–14:26 UTC, ≈ 77 by evening A local training engine that watches your coding agent's conversations and, while you're idle, distills lessons into a LoRA adapter for your own model. Runs entirely on your machine with version checks and rollback. Full card: https://gitnova.dev/en/r/sous-labs/songe.md 5. **microsoft/markitdown** — 4.7 · Peaking · Language models · Python · +276 stars measured 2026-11-22, 00:10–14:26 UTC, ≈ 446 by evening A Python utility from Microsoft for converting PDFs, Office documents, images, audio, HTML, and other formats to Markdown, aimed at LLM and text-analysis pipelines. It preserves document structure (headings, lists, tables, links) and is… Full card: https://gitnova.dev/en/r/microsoft/markitdown.md --- Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score. Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast. Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself. Data as of 2026-11-22 14:35 UTC, updated every 30 minutes.