AI tool intelligence

Find the right AI products for the work.

Our live catalog contains 2,444 researched AI products. Each record keeps current product positioning, fit, limitations, pricing signals, and official destinations together for a clearer decision.

Search products, capabilities, and workflowsBrowse the index ↓
2,444AI products in the catalog
DatedAvailability and evidence checks
IndependentFit, limitations, and alternatives

Public AI tool catalog

Search this public selection by the work you need to improve.

Search the complete catalog by product, capability, workflow, category, or tag. Every result opens a profile built from the same live record.

LA
Catalog evaluation

Data labeling

Labelbox

Labelbox is a platform and labeling service for creating high-quality training and evaluation data for AI systems. Teams can manage multimodal datasets, configure annotation projects, use expert labelers, and run workflows for RLHF, model evaluation, fine-tuning, and agent tasks.

Training dataRLHFModel evaluationMultimodal dataHuman feedback
Review
SA
Catalog evaluation

Data labeling

Scale AI

Scale AI provides training data, model evaluations, and full-stack AI systems for enterprises, governments, and model developers. Its products support data annotation and management, generative AI application development, red-teaming, benchmarking, and the deployment of AI systems with human review.

Data annotationHuman-in-the-loopRLHFModel evaluationGenerative AI
Review
SU
Catalog evaluation

Data labeling

SuperAnnotate

SuperAnnotate is a data platform for creating, curating, annotating, and evaluating multimodal training data. It combines customizable annotation workflows, quality review, managed expert talent, and integrations to support AI model training, fine-tuning, reinforcement learning, RAG evaluation, and agent development.

Multimodal dataRLHFFine-tuningRAG evaluationHuman-in-the-loop
Review

How the index works

A catalog built for decisions, not rankings.

Commercial relationships can fund the research. They never change product status, fit, limitations, evidence, or the alternatives we show.

Verify what exists

We check availability, product direction, pricing signals, and when the supporting evidence was last reviewed.

Evaluate operational fit

We assess the work it supports, integration requirements, ownership, risk, and where human judgment remains necessary.

Keep the comparison honest

Commercial relationships are disclosed, while credible limitations and alternatives stay visible.

Evaluating a product for a live workflow?

Bring the decision to our engineers