Cainew

LiquidAI/LFM2.5-230M is a lightweight foundation model with 230 million parameters designed for efficient inference and deployment. It offers a balance between model capability and computational resource requirements.

unsloth/Qwen-AgentWorld-35B-A3B-GGUF is an optimized 35-billion parameter agent model in GGUF format for quantized inference. It combines Qwen's capabilities with agent-focused architecture for efficient multi-agent applications.

deepreinforce-ai/Ornith-1.0-35B-GGUF is a 35-billion parameter reinforcement learning-focused model available in GGUF quantized format. It targets decision-making and learning tasks with optimized performance for resource-constrained environments.

Most AI teams pick a model first and discover the bill later. We built Oxlo.ai to change that. Access 35+ frontier AI models including DeepSeek V4 Pro, Kimi K2.6, GLM 5, Qwen, Llama, and Mistral through a single API. Compare models, calibrate responses, and choose the right model for each use case. Scale across AI models with predictable monthly subscriptions, benchmark-grade performance, generous usage limits, and we never train on your data.

BrowserAct is built for agents using the web. It gives agents a browser layer for real websites, so they can pass blocked pages, adapt to real scenarios, run multiple tasks safely, and return clean web data for reasoning. Use BrowserAct when an agent needs to browse, click, extract, fill forms, upload files, work inside logged-in sites, handle verification, or run repeatable browser workflows.

Zaro is where you can build working software from your scattered work. Everything you know is spread across Gmail, Slack, notes, and tabs that don't talk - Zaro pulls it into one place and lets you build apps from it in minutes: your research, your side projects, your plans, your decisions. Then they keep themselves updated, checking your connections every day so you don't have to. No code. No maintenance. No graveyard of prototypes you started and never finished.

Brain² knows every task, Doc, chat, and decision, and how they all connect. No re-explaining required. Pick any AI model, connect all your apps, and let AI work with full context of your company. Memory, agents, and MCP integrations built in.

Built specifically for the best product leaders, Samepage Signals automatically surfaces the information and insights that are most important to you across all of your tools and web, in one place.

Create Live AI teammates for every tough sales conversation. Each teammate shows up with human-like voice, avatar, and tools in minutes, can be deployed to phone, Google Meet, Zoom, or your application, and self-improves with every conversation. Use them to qualify leads, run autonomous product demos, coach new sales reps, and support your team live when the deal is on the line.

Grass gives your coding agents their own always-on cloud computer. Run Claude Code, Codex, or OpenCode on Grass, then monitor progress, approve decisions, and push changes from your iPhone. Version 2 is faster, redesigned, and now live on the App Store.

AI coding agents are limited to how autonomously they can work because they have no model of the codebase as a whole. Polygraph is a meta-harness that gives agents what they're missing: Visibility across every repo boundary and memory that survives the session. Connect all your repos, private and public, into a unified dependency graph without moving any code. Resume, reference or build on any session created by any developer, on another machine, even on different agents.

A developer created a Google Trends-style tool specifically for Hacker News that analyzes and visualizes trends from 18 years of comments, allowing users to track topic popularity over time on the platform.

A project implementing the Bible as a Retrieval-Augmented Generation (RAG) database, enabling users to query and retrieve biblical passages using AI-powered search and context.

This tool converts native language audio into interactive flashcards and provides shadowing practice to help users learn languages more effectively through dual input methods.

An article exploring techniques for combining visual and textual code representations to improve code understanding, documentation, and development workflows.

While Video Virtual Try-on (VVT) has achieved remarkable progress in synthesizing realistic garment overlays on dynamic subjects, existing paradigms remains fundamentally constrained by a passive dependency on source camera trajectories, failing to accommodate the requisite interactive freedom for omnidirectional viewpoint exploration. To address this limitation, we define a pioneering research frontier: Camera-controllable Video Virtual Try-on (CaM-VVT). Unlike conventional VVT, CaM-VVT not onl...

Tool Calling and Structured Output are two core capabilities of modern Agent systems, yet their interaction under joint deployment conditions remains insufficiently understood. This paper reports a reproducible phenomenon observed in a production Agent system: when Tool Calling and JSON Schema constraints are simultaneously enabled, multiple open-weight models cease invoking tools despite maintaining high schema compliance. We refer to this behavior as Tool Suppression. Through controlled experi...

Real-world photography requires capture-time guidance for both camera framing and subject pose. Yet existing aesthetic cropping benchmarks mainly evaluate post-hoc crop prediction and overlook subject-side recommendations, leaving the capture-time guidance capabilities of multimodal large language models (MLLMs) underexplored. To address this gap, we introduce CaptureGuide-Bench, a benchmark with two complementary tasks: photographer-side composition decision and refinement, and subject-side sce...

Fine-grained visual reasoning requires multimodal large language models (MLLMs) to identify task-relevant visual evidence and ground their reasoning in local image regions. Existing agentic methods typically rely on reinforcement learning with verifiable rewards or supervised fine-tuning on large-scale annotated reasoning traces, leading to costly exploration, hand-designed verification rules, or heavy dependence on textual supervision. A natural way to avoid such external answer labels is to le...

Synthesizing a novel-view video from a monocular reference video along a target camera trajectory requires both geometric consistency and motion fidelity with respect to the reference video. Existing methods based on explicit 3D representations are limited by the accuracy of off-the-shelf reconstruction modules, which often produce inaccurate geometry for dynamic objects in monocular videos. In contrast, camera-conditioning-only methods can achieve high visual quality but often struggle to prese...

Autoregressive video diffusion with causal diffusion transformers has emerged as a major paradigm for real-time streaming video generation and action-conditioned interactive world models. In this work, we extend rCM, an advanced diffusion distillation framework, to autoregressive video diffusion. The core philosophy of rCM lies in the complementarity between forward and reverse divergences, represented by consistency models (CMs) and distribution matching distillation (DMD), respectively, in dif...

Open domain subject-driven text-to-video (S2V) generation has drawn significant interest in academia and industry. Open domain S2V mainly involves two scenarios: in-domain, which requires retaining the reference subject features as much as possible, and cross-domain, which preserves the intrinsic features of the subject while allowing subject-irrelevant properties to vary flexibly according to the text prompt. Existing methods primarily focus on maximizing subject fidelity in in-domain scenarios...

We introduce Autodata, a general method that enables AI agents to act as data scientists who build high quality training and evaluation data. We show how to train (meta-optimize) such a data scientist agent, so that it learns to create even stronger data. We describe the overall formulation, and a specific practical implementation, Agentic Self-Instruct. We conduct experiments on computer science research tasks, legal reasoning tasks and reasoning with mathematical objects, where we obtain impro...

Modern large language models are predominantly trained with autoregressive factorization and causal attention. We present iLLaDA, an 8B masked diffusion language model trained from scratch with fully bidirectional attention. iLLaDA keeps the masked diffusion objective throughout pre-training and supervised fine-tuning (SFT), scaling pre-training to 12T tokens and fine-tuning on a 25B-token instruction corpus for 12 epochs. We further use variable-length generation for efficiency and introduce co...

Anthropic has accused Alibaba of illicitly extracting Claude AI model capabilities without authorization, raising concerns about intellectual property protection in the AI industry.

Ford has rehired experienced veteran inspectors after its AI quality control systems produced errors, highlighting limitations in fully automated inspection processes for manufacturing.

Medical students are misusing a popular research tool to generate and publish misleading studies, raising concerns about research integrity and the ease of producing fraudulent academic content.

The article discusses how open-weight AI models have become remarkably inexpensive, challenging proprietary model economics and democratizing access to powerful language models.

A new OpenAI research paper shows how AI agents are transforming work, enabling longer, more complex tasks and expanding productivity across roles.