Havuz

Pre-trained subword embeddings in 275 languages, based on Byte-Pair Encoding.

IdeaTaste - A benchmark for research taste: let LLM agents propose research ideas, then check how many of them the research area actually we

Редактура українського тексту: прибирає сліди ШІ, канцелярит, суржик і кальки, зберігаючи зміст і голос автора. 63 маркери, 30 заборон, скан

AIが書いた日本語を推敲するスキル yomiyasu を、スキルの大きさに上限がある環境向けに短くしたスキル（SKILL.md 18,000字以下）

Runs Hugging Face ONNX models on any Nx backend, with no per-architecture code

Xing4.0-29B-Intent - schema-guided intent recognition and structured information extraction (intent and slot filling, entity, relation and e

Train JEV-style System 1 decision models on Qwen3.5 and Qwen3.8-27B: 24-slot decision head, KL distillation, calibration and serving. 在 Qwen

Fold look-alike CJK glyphs (Kangxi radicals, CJK radical supplements) so text extracted from PDFs matches again.

🎙️ Real-Time Multi-Modal Speech Telemetry & Acoustic DSP Engine (FastAPI, Python 3.11, Autocorrelation Pitch, FFT Spectral Energy & Lexical

A curated corpus of Nigerian legal texts, including the 1999 Constitution and selected Acts, cleaned and structured for information retrieva

Offline linter that marks LLM-sounding prose. Its statistical model is trained on 2.8B tokens of human text and 11 LLM families, with 0.96 m

yomiyasu, Codex, Claude Code ve Cursor gibi agent ortamlarında AI tarafından üretilen Japonca metinleri daha doğal ve okunabilir kılmaya yönelik bir Agent Skill sunuyor.

Neden yükseliyor · Bir günlük repo yaşına karşılık 625 yıldız, dikkat çekici erken ilgiye işaret ediyor. Ancak bu ilginin kullanıcı benimsemesine dönüşüp dönüşmediği bilinmiyor; son 28 günde 2 commit ve 1 farklı yazar, bakım ve topluluk katkısı için henüz güçlü kanıt oluşturmuyor.

A decision model that returns calibrated answers to yes/no, choice, score and multi-label questions.

Classify text in PHP without an LLM bill: typed decisions in 100+ languages, self-hosted. Laravel-ready SDK for Laya, a Jev AI alternative.

Asking for What Was Never Requested: horizontal and vertical proactivity in LLM agents. Need-graph metrics (no LLM judge) and Q&D, which tra

日本語の文章から「AI 臭さ」を機械的に拾う Rust 製の Linter

Quirón, Claude Code, Codex ve Cursor için metni AI yazım kalıplarına göre düzenleyen ve değişikliği ölçen bir agent skill’idir.

Neden yükseliyor · Beş günlük bir repoda 49 yıldız ve son 28 günde 42 commit dikkat çekiyor. Ancak farklı yazar sayısının 1 olması, aktivitenin tek bir geliştiricide toplandığını gösteriyor; yükselişin nedeni veya kullanıcı benimsemesi hakkında daha güçlü kanıt yok.

Best Open Source NLP LLM Fine-Tuning Trainer Framework 2026

AdalFlow Alternative 2026: Auto-Optimize LLM Apps Without Code

VedAnvaya — an evidence-led digital atlas for reading, exploring, hearing, questioning, and navigating connections across the Vedic corpus.

Text extraction with Jev: field extraction, PII detection and exact quotes, built on TypeSafe's Jev.

Research project on bilingual Urdu-English retrieval-augmented generation for Pakistani government and legal documents, with OCR, hybrid ret

High-performance neural networks engine for web, built on top of TypeGPU

中国法律法规与案例结构化语料库：446 部法律/司法解释逐条 JSON（含效力状态）+ 278 指导性案例 + 445 公报案例，CC0

Hybrid rule-based and context-aware LLM evaluation system for Chinese-to-Japanese game localization QA.

Zero-cost epistemic uncertainty quantification & hallucination detection for LLMs (90,000x faster than Semantic Entropy)

Build & Auto-Optimize LLM Apps in 2026: The Ultimate AdalFlow Guide

Master LLM Fine-Tuning Trainer 2026: Advanced NLP Optimization Toolkit

Translate between English and Claudish with tiny ProgramAsWeights functions.

Wenshu（文枢）— 面向人文社科研究的 AI 知识处理工作流：本地知识库 / RAG / 中文引文 / 理论谱系。AI knowledge workflow for humanities & social sciences: local knowledge base, RAG

日本語文書の自然さと読みやすさを再現可能に診断するRust CLI / Deterministic diagnostics for natural and readable Japanese writing

Remove AI text watermarks (SynthID-Text, KGW, Unigram) by regenerating text from its meaning — scheme-blind, runs on local models, every cla

A consistent, fair, benchmark for structured extraction.

Prepackaged medical pronunciation dictionary for voice AI TTS engines. 960+ drug names, clinical terms, anatomical terms, and medical acrony

工厂本体驱动的数据问答框架（开源，Apache-2.0）：把中小企业台账/MES/ERP 变成可自然语言问答的语义知识图谱。CSV进答案出，确定性规则引擎+本体GraphRAG+向量混合检索，本地化数据不出厂，换行业只换词典。

Privacy and security guardrails for Spring AI: request-scoped PII tokenization, least-privilege disclosure to tools, and tool authorization

AI-powered fake news detection system with multiple ML models and explainability.

Let your users ask your database anything, in their own words — safely. The model fills in a validated Query AST, never a query string, and

Teknofest 2026 Doğal Dil Ajanları Yarışması 1. Senaryo kapsamında geliştirilen, kamu sektörü için resmi evrak analizi ve taslak üretimi yapa

Local, per-sentence AI-text detection with pattern diagnostics — CLI + MCP server (DeBERTa-v3, RAID #1).

Hybrid recommender system + RAG chatbot for IT courses (FastAPI + Express/TS + PostgreSQL), with an independent Data Lakehouse module (Spark

SomNLP-Translate is an open, high-quality English–Somali parallel dataset designed to support research and development in machine translatio

A mental-model-first field guide to Audio AI — from raw waveforms to production voice systems. Covers ASR, TTS, audio classification, Transf

Multilingual readability engine for Node.js/TypeScript - 86 languages, 17 formulas, TeX (Frank M. Liang) syllable counting. Zero runtime dep

Python RAG-based chatbot for document Q&A using retrieval-augmented generation

natural-japanese, AI agent’ların iş amaçlı Japonca metinleri daha okunaklı yazıp gözden geçirmesine yardımcı olan bir Agent Skill ve lint araçları sunar.

Neden yükseliyor · 78 günde 1.748 yıldıza ulaşması, AI ile üretilen Japonca metinlerin niteliğine dönük ilginin güçlü olduğuna işaret ediyor; birden fazla agent ortamında kurulabilmesi de erişimini genişletiyor. Ancak bu ölçümler ilginin nedenini kanıtlamıyor; son 28 günde 5 commit ve 1 farklı yazar görülmesi de geliştirme faaliyetinin dar bir katkıcı çevresinde olduğunu gösteriyor.

Multilingual readability library for Rust - 86 languages, 17 formulas, TeX-based syllable counting via the Frank M. Liang algorithm. Byte-ac

Python SDK for LLM guardrails with safety classification, PII detection, prompt injection defense, and grounding checks. Protect AI applicat

The fastest lightweight, embeddable tiktoken-compatible Rust core.

A free, privacy-first tool that flags the tells of AI-generated writing, overused vocabulary, rhetorical crutches, robotic sentence rhythm a