Home

Open Source

Curated open-source AI projects with verified GitHub data and hands-on breakdowns.

Univer Office SDK: Agents Edit Cells Without an AI SDK

Only what the dream-num/univer README, LICENSE and GitHub API fields support. Apache-2.0 covers the Web SDK core architecture and engine implementations; commercial plugins, the Server SDK and commercial AI SDK extensions are distributed under separate licences. Facade APIs such as FWorkbook, FWorksheet, FRange, getRange and setValues let an application or Agents Edit Cells directly with formula recalculation, while workbook.save() persists a JSON snapshot rather than XLSX import or export. No AI SDK is required anywhere in that path. Stars are cumulative GitHub API totals, not single-day growth.

llama.cpp and ggml Ship v1.0.0: Deprecated APIs Removed

Reads only what the GitHub API fields, the two v1.0.0 release notes, Discussion 1579, LICENSE and README of the ggml-org llama.cpp and ggml repositories can verify. Both v1.0.0 tags landed on November 13, 2026. It separates semantic versioning, the rule that a release tag requires the internal ggml copy to match the released one, the split between b-numbered nightlies and vX.Y.Z releases, and the 35 removed functions, RPC v9 and MIT notice-retention and AS IS terms. The August 2026 versioning discussion is not presented as a November announcement, and no figure from the unretrieved HF blog is quoted.

LingBot-Touch: Code Apache-2.0, Weights Elsewhere

Robbyant/lingbot-touch reached 238 stars within days of creation, with Apache-2.0 in both the license field and the badge alt text, yet the model weights are not in the source repository: the README states verbatim that model files are hosted separately and access requires an account with permission to the model repository. This piece reads only API fields, README wording, and the LICENSE text: created November 9, 2026, last push November 11, main language Python, default branch main, not a fork and not archived. It also covers how three heterogeneous tactile sensors feed one representation and how the two model variants differ, without importing scores, download counts, or star totals the repository never states. As of November 12, 2026, code licensing and gated weights remain two separate questions.

16 Megapixels, No Upscaler: What Does the Repo Actually Ship?

MiniMax H3-Image is an open-weight image generation model whose official repository MiniMax-AI/MiniMax-H3-Image (hyphenated GitHub organization) holds, as verified on November 11, 2026, no runnable inference code - only a model card, a license and resource files, with language unset and a single commit. The README model card states 33.1B parameters in a dense single-stream transformer, distilled to 6 sampling steps (6 NFEs) with no classifier-free guidance, about 16 megapixels at every aspect ratio directly and without an upscaler, BF16 DiT weights at 68.65 GB, and a Qwen3-VL text encoder kept outside the repository; the official launch post separately claims a 4K (4096 x 4096) image in under 10 seconds, Standard 1.5-2x and Highspeed 3-4x speeds, pricing at 50-70% of mainstream offerings and a launch-week 50% off for 4K-and-above requests - all launch-post wording, with no seconds figure anywhere in the README. The API guide caps output area at 20,000,000 pixels, allows 1 to 16 reference images and 1 to 10 images per request, and both endpoints are synchronous. The license stays a one-sentence boundary: community license with GitHub NOASSERTION, terms per the in-repo LICENSE, angled against the October 24 M3.1-Flash deployment SOP.

An Agent That Lives on Your Phone: Inside the 7,938-Star Mika

wangzan101/Mika is an open-source on-device agent that installs on your Android phone and does the tapping, typing and scrolling itself: a GitHub API snapshot on November 10, 2026 shows 7,938 stars, Kotlin as the primary language, an Apache-2.0 license, and a last push at 04:19 UTC the same day; the November 7, 2026 metadata-review snapshot in our archives counted 7,489 stars, a net gain of 449 with both timestamps attached. The README positioning is blunt: you say what you need, and it works your apps and the web until it is done; the project has a homepage at mikaphone.com and multilingual READMEs. The deep dive covers the background-screen mechanism, Ask and Autonomous modes, three model connection paths, Android 11/12/14 requirements, a built-in app knowledge table, a three-layer privacy boundary, Releases-based installation and a 12-item compatibility self-check, all checked against the README text; unspecified architecture, roadmap and author details are left out rather than invented.

Announced GA, Yet Only 71 Stars: What Is Kinference?

Buoyant, the company behind Linkerd, open-sourced Kinference as an inference control plane for Kubernetes and vLLM and announced GA on November 4, 2026; a GitHub API snapshot on November 9, 2026 shows Kinference-AI/kinference (migrated from buoyantio/kinference via 301) at only 71 stars, 6 forks and 57 open issues, in Rust under Apache-2.0, created October 5, last pushed November 8, version v0.1.0-alpha.7. The README confesses verbatim that Kinference is alpha software with no sustained production exposure, lacking Anthropic-style /v1/messages, rate limiting, semantic classifiers and multi-replica support. Mechanically it exposes a gRPC interface that takes OpenAI-format requests, returns the chosen endpoint plus a mutated body, monitors vLLM KV-cache events and engine metrics, reuses NVIDIA's dynamo-kv-router crate and pairs with the Buoyant Gateway. Its vendor-reported median TTFT improvement, 6.88s to 0.31s on shared-prefix traffic, was measured against random routing rather than llm-d or Dynamo, and has no third-party reproduction.

3,003 Stars, No Stable Release Yet: What Is Blocking It?

oxc-project/tsrs is a Rust rewrite of the TypeScript 7 compiler and language server, described in its README as a drop-in replacement. A GitHub API snapshot on November 8, 2026 showed 3,003 stars and Apache-2.0, but still no stable release. The README says the team aims to ship a stable version against TypeScript 7.1 with 100% conformance; aim is a target, not a delivered result. docs/STATUS.md lists GitHub prereleases, a manually assigned npm beta tag and release candidates, but nowhere declares stable. Speedup figures come from three unrelated setups that cannot be merged into one table: the project's own bench/RESULTS.md covers 19 projects on an Apple M3 Max, measured 2026-10-28 with Hyperfine medians against TypeScript 7.1.0-beta, giving 1.71x-6.62x faster; Cloudflare's up to 8x compares against TypeScript 7.0; third-party numbers near 2.7x and 5.3x have their own machines and baselines. This is a deep dive into one repository, not a multi-project ranking.

19,207 stars, still alpha: can this Rust engine rebuild a site?

storytold/craft-anything describes itself as an open-source clean-room engine in pure Rust: see how any website, app or game works, then have AI build your own version. The workflow points at a target, states what is wanted, studies the original with real inspection tools, has a coding agent build a new project, compares results against the original and iterates. A November 7, 2026 GitHub API check showed 19,207 stars, Rust as the language and a repository created on October 31, 2026. The README badges say early alpha and 100% Rust, and the license text is dual-licensed under MIT or Apache-2.0 at your option. ATTRIBUTION.md records bundled font and icon licenses while brand assets must not be modified. macOS, Windows, Linux, FreeBSD and Web are supported through a desktop app, CLI and MCP server, with Anthropic, OpenAI, OpenRouter, xAI and local models plus installed GitHub Copilot, ACP grok and Ghidra. The README admits some capabilities are built but not yet run live.