Alternatives to Label Sleuth

Free browser tool from Google for training image, sound, and pose classifiers with no code and no account.

Open-source data labeling and AI evaluation across every data modality, self-hosted.

Load, stream, and preprocess training datasets in one line of Python — open source, Arrow-backed, Hub-native

CVAT is an open-source data annotation platform for image, video, 3D point cloud, and audio labeling.

Open-source data annotation for text, image, audio, video, 3D, and AI agent traces — built in YAML, self-hosted free.

Flower runs federated learning across decentralized, private data with the open-source Flwr framework and SuperGrid.

Prodigy is a downloadable Python annotation tool for building training and evaluation data for custom AI, ML and NLP models.

Browser and offline desktop data labeling for image, text, document, and video annotation.

Open-source, project-based tool for turning your documents into QA datasets for LLM fine-tuning.

DataQA turns Slack into a weak-supervision console for labeling text.

Mundo AI builds culturally-aware audio and video training data, evaluations, and applied research for teams working on perceptual AI.

Custom audio datasets for training and evaluating voice AI, scoped to your model's spec.

Clickworker turns a verified global crowd of 10M+ people into on-demand data collection, labeling, and validation work for AI, retail, and research teams.

Mercor connects AI labs with vetted domain experts to build training data, evals, and APEX benchmarks for frontier models.

LabelGPT auto-labels images from a text prompt—zero-shot bounding boxes and segmentation in minutes.

Prolific connects AI and research teams to 300,000+ verified participants for human feedback, preference tuning, and safety evals.

Klavis AI builds live Dockerized environments for training AI agents on coding and real tool-use tasks.

Training data, RL environments, and red-teaming for LLM and computer-use agent teams

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

Synthetic data generation for computer vision that ships photorealistic, auto-annotated training sets on demand.

Licensed conversational audio datasets for training speech and omni-models, with samples in 48 hours.

Generate LLM training datasets with DAG topic graphs and real sandboxed tool-calling traces

Snorkel AI builds expert-curated training data and runnable eval environments that fix where frontier models and agents break.

Managed human data annotation for computer vision and NLP — bounding boxes, polygons, cuboids, keypoints, segmentation, text and audio labels, priced per

Hub recruits the physical world to capture egocentric, teleoperation, hand-tracking and tactile training data for embodied AI.

Markup turns PDFs and long text into structured NLP and ML datasets with AI-assisted annotation in the browser

Enterprise dataset management platform for AI training teams

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

Applied AI for aviation: pilot-controller training datasets, the Y4 Aviation Understanding Model, and real-time runway incursion risk detection.

Liva AI builds human interaction datasets and social-intelligence evals that help models read intent, emotion, and context.