Alternatives to Markov

Offline AI image annotation for YOLO and VOC datasets — local SAM2, SAM3 and YOLO-World, one-time purchase from $49.

Applied research lab that captures expert reasoning and structures it into SFT, RL rubric, agent, and computer-use training data for frontier models.

A human data platform connecting AI teams and researchers to 300,000+ verified participants for preference tuning, safety evals, and online studies.

GameEngineBench benchmarks and UE5 training data for AI agents that write real game-engine code.

Velvet sources, verifies, and delivers curated video datasets that help AI labs train interactive and world models with real spatial reasoning.

Snorkel AI builds expert training data, evals, and runnable environments for frontier models and agents.

Open-source Python framework for generating grounded synthetic datasets from real sandboxed tool execution traces.

BioStack Platforms supplies structured clinical datasets and RL environments for healthcare AI teams.

Open-source, multimodal data annotation platform for image, video, 3D point cloud, and audio labeling—self-hosted, cloud, or fully managed

Free, open-source desktop AI assistant that runs 24/7 on your own machine, scraping and analyzing web sources autonomously.

Allyhub is an AI agent platform that automates web browsing, research, and reporting with no code.

Cortex AI supplies real-workplace egocentric video and robot trajectory data for training embodied AI models.

Early-stage San Francisco robotics lab building policies for robots that do real work, and hiring its first researchers.

A free 70,000-image Zalando clothing dataset built as a drop-in replacement for MNIST.

Open-source, self-hosted annotation for text, media, and AI agent traces — configured entirely in YAML.

Raiinmaker builds custom, rights-cleared video datasets and runs real-time human evaluation for generative AI video models.

Autoscaled RL data and self-evolving benchmarks for tool-use and computer-use agents, built by the team behind Ragas.

Multimodal real-world human data for training embodied AI agents

Spider Cloud is a web scraping and crawling API that turns live pages into markdown or JSON for agents and RAG pipelines.

OpenAgents is an Apache-2.0 platform for language agents that analyze data, call 200+ plugins and browse the web.

Open-source MCP server that gives coding agents live Chrome DevTools access for debugging, automation, and performance traces.

Adept builds AI agents that chain real actions across desktop and web apps, adapting when interfaces change.

Firecrawl is a web data API that turns any site into clean Markdown, structured JSON, or screenshots for AI agents.

Discontinued: Gobii's AI recruiting agents wound down, with the final day of operations on September 1, 2026.

Cloud browser sessions, sandboxes, and AI agents for web scraping and agentic automation.

Open-source platform for data labeling, AI evaluation, and human-in-the-loop workflows across every data modality.

RL environments and expert-labeled data for frontier models and enterprise agents

Managed training data, RL environments, and safety red-teaming for teams building AI agents, coding models, and LLM/VLM pipelines.

Open-source SDK for building browser agents with AI primitives (Act, Extract, Observe, Agent) plus Playwright-style controls in TypeScript, Python, and Go.

ClawBench benchmarks AI browser agents on live websites with HTTP-interception scoring and LLM-judge grading.