Havuz

An experiment-driven laboratory for understanding LLM sampling strategies. Explore how Temperature, Top-k, and Top-p influence token probabi

A recursive self-improvement (RSI) workflow that evolves both target agents (e.g., Pi and DeepSeek Harness) and the RSI system itself.

Vela: User Behavior as Supervision Sparks LLM Intelligence — official paper, task templates, and prompts from AllSpark at Xiaohongshu.

Official implementation of “ReSAIL: Mitigating Collapse in Iterative Agent Self-Distillation”

STEPQuant: When and Where Errors Matter in Delta-Rule Recurrent State Quantization

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

Myrtia — agent-native red teaming for LLMs and AI agents: a deliberating council of attacker models (propose, cross-critique, Borda vote, ch

AI Model World, yüzlerce AI modelini fiyat, yetenek, benchmark ve yayın tarihi verileriyle karşılaştıran piksel-art görselli bir web sitesidir.

Neden yükseliyor · Güncel model karşılaştırması ve görsel sunum, hızla değişen AI ekosisteminde ilgi çekici bir kullanım sunuyor. 11 günde 191 yıldız ve son 28 günde 56 commit dikkat çekici bir başlangıç; ancak iki farklı yazar bulunması ve kısa geçmiş, benimsemenin kapsamı ya da sürdürülebilirliği hakkında henüz güçlü kanıt oluşturmuyor.

Live Benchmark for long Horizon tool using Agentic Systems

A clinical decision-evidence benchmark: pull the one fact an answer rests on, and see whether the assistant's action moves with the evidence

NeoHorse, agent iş akışlarında tool use, coding ve yapılandırılmış kararlar için açık ağırlıklı 4B ve 9B modeller sunuyor.

Neden yükseliyor · 23 günde 1.094 yıldıza ulaşmış; son 28 günde 13 commit ve 6 farklı yazar katkısı da projede yakın dönem geliştirme hareketi olduğunu gösteriyor. README'de model yayını ve arXiv teknik raporu duyurulmuş; ancak ilgi ve ekip aktivitesi, modellerin gerçek kullanımını veya bağımsız başarı teyidini tek başına kanıtlamıyor.

The best model under 200B at near 10 token/s exact batch throughput on a single laptop CPU, with an automatic 8 GB RAM path. Native C, no GP

Halo is an open-source framework built by White Circle for training large language and multimodal models

Evaan — a fully local, offline-capable AI companion chatbot built with Python & Hugging Face Transformers, running on Qwen2.5-0.5B-Instruct

Qwen3.8-27B on one laptop CPU: up to 2.52 token/s, 8 GB tested, no runtime accuracy loss. Native OpenAI-compatible API with function tools;

Run the native 284B-A13B DeepSeek-V4-Flash-0731 LLM locally on one laptop CPU: pure C, 8 GB RAM minimum, no GPU, best TPOT 0.892 s/token, re

Minimal harness-agnostic recursive language model layer — one binary, Python + llm()

An end-to-end framework for Agentic Reinforcement Learning.

Benchmark for repository-level code understanding: 500 open-ended questions from real GitHub Discussions across 50 repositories, graded by a

A full-screen terminal environment for writing fiction with language models.

An agentic LLM-powered knowledge assistant that enhances RAG capabilities through automated entity extraction, structured data analysis, and

🚀 One API gateway for GPT, Claude, Gemini, DeepSeek and more. OpenAI-compatible API for developers.

Tools Are Not Islands: Set-Level Tool Retrieval for LLM Agents via Query-Conditioned Hyperedge Prediction

大模型故障 token（glitch token）资料库：收录各 LLM 分词器中已公开记录的异常 token，附异常表现、来源与复现链接。A curated catalog of LLM glitch tokens.

670 MB/s Swift tokenizer inspired by Marcel Roed’s gigatoken. Zero-copy, NEON/CRC32, WebAssembly, and Embedded Swift.

hummingbird is a lightweight, zero dependency runtime for massive open source Mixture of Experts (MoE) language models. It unifies SSD, RAM,

Benchmarking data agents for verifiable analytics over heterogeneous workspaces.

GLM-5.2 NVIDIA NIM Go 客户端 / OpenAI 兼容反向代理 — 自动化 hCaptcha 凭证池 + 流式推理，支持 Docker 部署。Go client & OpenAI-compatible proxy for NVIDIA Playground's

Interactive 3D visualization platform for exploring transformer architectures, tensors, and real-time LLM inference.

An intentionally vulnerable OWASP LLM Top 10 training platform for AI Security, Prompt Injection, RAG Security, Agent Security, and GenAI pe

GLM-5.2, a 744 billion parameter mixture of experts model, in a pure C inference engine: quantized to int4, experts streamed from disk, depl

Jacobian-Brainwash : A manual alignment tool for large language models built on Anthropic's Jacobian Lens. Results are exportable.

Take the 100 Days of AI challenge - learn one AI concept a day, from machine learning basics to AI agents in production, in 100 days.

Transformer architecture explained step by step - the full architecture, every attention variant, positional embeddings, and every layer ins

A practical Generative AI engineering lab exploring LLM foundations, tokenization, deployment strategies, and real-world AI system design.

🧠 One LLM, split across a Mac (Apple MPS) + a Windows PC (NVIDIA CUDA) — heterogeneous pipeline inference over a framework-neutral wire, bit

Simulate AI Personas in Social Feeds with Real-Time Chat Bots 2026

Daily LLM value rankings - compare 300+ models by intelligence, speed and price. OpenRouter + Artificial Analysis. 大模型性价比排行榜

从 MiniMind 源码读起,再延伸到现代大模型技术体系的中文学习笔记。主线逐行精读预训练 / SFT / DPO / PPO / GRPO 与训练机制;附录 17 篇进阶卷覆盖量化、投机解码、RLHF 全景、模型代际史等 MiniMind 没涉及、但进阶绕不开的主题。

爻枢 Token 智能网关、Token 智能聚合、大模型评测、排名等。关注“爻枢未来”公众号获得更多全自动编程详情。

🌟 [Survey & Benchmark] A unified taxonomy, geometric framework, and cross-domain benchmark for modern optimizers

Seven Minutes Is All I Can Spare To Play With You.

Source code of paper "RLCSD: Reinforcement Learning with Contrastive On-Policy Self-Distillation"

AI 创始人 & 研究员长访谈深度分析库 | 张小珺商业访谈录、WhynotTV、雪球《方略》结构化拆解：段永平、姚顺宇、翁家翌、何小鹏、杨植麟(Kimi)、罗福莉、张鹏(智谱)

Autonomous web-app pentest engine — a multi-agent pipeline wielding 30+ offensive-security tools as skills, PoC-validated findings. Built on

Desktop provider manager and local failover router for Codex, Claude Code, and Grok.

Molecode presents molecules as code and enables LLMs to operate and reason on chemistry directly.

211 free visual cards about large language models, from tokenization to deployment. Full collection: 376 cards.

面向智能座舱的云边协同 AI Agent：端侧毫秒级车控，云端声明式 Multi-Agent 与 Skill/DAG 编排，支持 S2S 实时语音、声纹多用户、HMI和Android双端；LLM 只负责理解与规划，VAL 负责确定性安全执行。

Deterministic causality engine for AI roleplay.