Coactive turns video and image libraries into structured, searchable intelligence at petabyte scale. Using multimodal AI, it auto-tags content, enables semantic search, and chains models for deeper insights — helping media, entertainment, and enterprise teams manage, monetize, and measure unstructured visual data. The platform integrates with Bedrock, Azure AI, and Databricks.

AI-powered platform for extracting insights from any link - videos, articles, or PDFs. Features deep research with 4-layer Mixture-of-Agents, mind maps, research slides, and multi-model debate arena.

AI-powered global news analysis platform that automatically identifies patterns, themes, and emerging trends across worldwide news sources. Provides daily insights and trend reports without manual curation.

AI automation tool for scientific research that automatically generates literature reviews, finds relevant papers, extracts key findings, and creates structured research reports with citations.

Exa's new agentic search tier that delivers synthesized, reasoning-grade web results for AI agents and LLMs. Exa Deep offers two modes—'deep' (4-12s latency) and 'deep-reasoning' (12-50s latency)—designed to give AI agents comprehensive, semantically-processed web information rather than raw links. Launched March 2026 alongside an updated pricing structure.

The Isomorphic Labs Drug Design Engine (IsoDDE) is an AI system that moves beyond protein structure prediction into active drug design, achieving double the accuracy of AlphaFold 3 on complex structural predictions. Published in February 2026, it predicts binding affinity and identifies cryptic binding sites from just protein sequences, dramatically accelerating the drug discovery process from years to weeks.

Listen Labs is an AI-powered qualitative research platform that conducts deep one-on-one interviews at scale across 30M+ participants. It uses AI moderation, emotion detection, and dynamic probing to deliver consultant-quality analysis in under 24 hours. The platform raised $69M in Series B funding in March 2026 to expand AI-driven customer insights.

Collaborative data analysis platform launched in January 2026 that replaces traditional data tools like Jupyter, Hex, and Deepnote. Features Python, SQL, DuckDB, and Polars support with natural language querying.

Multi-agent AI research platform that conducts comprehensive research analysis across multiple sources. Uses coordinated AI agents to gather, analyze, and synthesize information for deep research tasks.

MAI Playground is Microsoft's AI model testing platform giving developers direct access to its latest in-house foundational models, including MAI-Image-2 (top-3 text-to-image on the Arena.ai leaderboard), MAI-Voice-1 (ultra-fast voice generation at 60 seconds of audio per second), and MAI-Transcribe-1 (high-accuracy speech-to-text). All models are available for commercial use via Microsoft Foundry and the MAI Playground interface.

Memories.ai is building the visual memory layer for wearables and robotics, partnering with NVIDIA to index and retrieve video-recorded memories for physical AI systems. Founded by former Meta researchers, it provides a large visual memory model that processes and retains visual data from device sensors, creating a searchable record of what AI systems have 'seen'. Designed to give wearables and robots persistent, recall-capable memory in the real world.

MiroThinker 72B is a powerful open-source reasoning model by Miro Lab that uses interactive scaling — running internal verification cycles before producing output rather than generating immediately. It achieves 81.9% on the GAIA benchmark, matching paid frontier model performance on complex logical reasoning tasks, and is available for free download and local deployment.

Personal AI radar that monitors thousands of sources 24/7 for topics you care about. Create custom trackers using natural language, get AI-summarized insights with source citations, and refine results through chat.

AI-powered knowledge threading platform that enables fast information discovery across all your data sources. Features a RAG API for developers to build agentic AI applications with enterprise data, supporting workflow automation and intelligent search.

A free AI-powered workspace for scientific writing and collaboration. It brings drafting, revision, collaboration, and publication into a single cloud-based LaTeX-native workspace with GPT-5.2 integrated directly into the authoring environment.

OpenHunt is an AI-native product launch and discovery platform where multiple autonomous agents analyze every submitted product and generate structured insights before the community even sees it. It's built for the post-algorithm internet, giving builders a smarter alternative to traditional launch platforms. Community members then vote on what truly deserves attention, backed by agent-generated signal.

AI research platform that searches millions of academic papers, provides AI-assisted Q&A with papers, network analysis for literature discovery, and interactive note-taking for researchers.

All-in-one AI research platform combining literature discovery, AI-powered review, systematic research synthesis, citation management, and PDF analysis for academic writing.

Perplexity AI's upgraded autonomous research agent that performs dozens of web searches, reads hundreds of sources, and delivers a structured, cited research report in 2–4 minutes. Launched March 15, 2026, it ranks among the best AI research tools for comprehensive multi-source synthesis.

Open-source embedding models for AI search systems, designed to improve retrieval accuracy and support multilingual search tasks with efficient memory usage.

Perplexity Model Council, launched February 5, 2026, lets users simultaneously query and compare outputs from multiple frontier LLMs — including GPT-5.4, Claude Opus 4.6, and Gemini 3 Pro — side by side in one interface. It helps users evaluate nuanced differences in reasoning, tone, and accuracy across models without switching platforms, available to Perplexity Pro subscribers.

Perplexity Pro offers unlimited usage of GPT-4, Claude 3 Opus, and other top models for deep research and web reasoning.

Open-source Retrieval-Augmented Generation engine that combines deep document understanding with agentic capabilities to build production-grade chatbots and knowledge bases. Supports PDFs, Word, slides, images, and 30+ data sources with grounded citations and minimal hallucinations.

ResearchPal is an AI research assistant for academics that integrates literature discovery, literature review generation, in-text citation management, Zotero integration, and academic writing all in one workspace. It helps students, researchers, and professionals accelerate thesis writing and research workflows.