Evals is an agent-experience evaluation platform scoring how product flows work for AI agents.
Evals lets devtool teams measure Agent Experience on any product flow using their own custom prompts and API keys, surfacing exactly where agents get stuck such as authentication issues, missing documentation, or confusing errors.
Teams run evaluations in staging to receive regression alerts before changes reach production, turning agent-interaction traces into prioritized fixes across the product surfaces that agents rely on.
| Attribute | Evals | AI Evaluation and Annotation Platform | Arize AX, Arize Phoenix | |
|---|---|---|---|---|
| Provider |  |  |  | |
| Founded | 2026 | 2025 | 2020 | |
| Sells To | B2B, Developer | Enterprise | B2B, Enterprise | |
| Pricing Model | B2B Subscription | SaaS | Recurring, SaaS, Subscription | |
| Ownership | Private | Private | Venture-backed |
Provider
Founded 2026
Sells To B2B, Developer
Pricing Model B2B Subscription
Ownership Private
Provider
Founded 2025
Sells To Enterprise
Pricing Model SaaS
Ownership Private
Provider
Founded 2020
Sells To B2B, Enterprise
Pricing Model Recurring, SaaS, Subscription
Ownership Venture-backed
StableBrowse Agent Experience Platform helps devtool, infrastructure, and SaaS companies make their products usable by AI agents.
Eval Engineering Platform turns AI evaluation results into automated production guardrails.
Analytics that ranks AI agent journey failures by business impact and tests fixes.
Synthesizes end-to-end agentic workflows grounded in real business logic from your estate.
AI agent that scans customer reviews, chats, and calls to surface insights and sentiment trends.
Advisory and benchmarking service using automated metrics and MQM-based human audits to validate engine performance and optimize MT/LLM workflows.
UserEvidence Evidence is a customer-evidence module for collecting, curating, sharing, and measuring verified customer proof.
An AI site reliability engineer that verifies production changes, investigates regressions, and proposes root-cause fixes.
Selfr's analytics agent answers structured-data questions using a non-autoregressive model built not to hallucinate.
Open source AI gateway routing every model behind one zero-markup OpenAI-compatible endpoint.
CodeGPT Goal-Oriented Agents manage customer journeys across chat, voice, email, and SMS channels.
Simple AI deploys voice, chat, text, and email agents that handle sales, support, and appointment booking.
StableBrowse Agent Experience Platform helps devtool, infrastructure, and SaaS companies make their products usable by AI agents.
Eval Engineering Platform turns AI evaluation results into automated production guardrails.
Analytics that ranks AI agent journey failures by business impact and tests fixes.
Synthesizes end-to-end agentic workflows grounded in real business logic from your estate.
AI agent that scans customer reviews, chats, and calls to surface insights and sentiment trends.
Advisory and benchmarking service using automated metrics and MQM-based human audits to validate engine performance and optimize MT/LLM workflows.
Adaptive OpenAI-compatible endpoint coordinating open-weight models for chat and agents.
UserEvidence Evidence is a customer-evidence module for collecting, curating, sharing, and measuring verified customer proof.
An AI site reliability engineer that verifies production changes, investigates regressions, and proposes root-cause fixes.
Selfr's analytics agent answers structured-data questions using a non-autoregressive model built not to hallucinate.
Open source AI gateway routing every model behind one zero-markup OpenAI-compatible endpoint.
CodeGPT Goal-Oriented Agents manage customer journeys across chat, voice, email, and SMS channels.
Simple AI deploys voice, chat, text, and email agents that handle sales, support, and appointment booking.
