專項更強簡化從標註到模型部署的端到端電腦視覺工作流程,但缺乏 Labelbox 的一般企業代理編排功能。
Labelbox 每月有 762,803 次造訪,月增率 -22.8%,定價模式為需聯絡報價。Unlearn.AI 每月有 17,107 次造訪,成長率 -32.5%,定價模式為需聯絡報價。

Unlearn.AI builds AI-generated digital twins of clinical trial participants — forecasts of how each individual patient's outcomes would evolve — and uses them to sharpen design, interim, and readout decisions in drug development.
| 面向/指標 | ![]() | ![]() |
|---|---|---|
| 官方網域 | labelbox.com | unlearn.ai |
| 定價模式 | 需聯絡報價 | 需聯絡報價 |
| 每月造訪次數 | 76.28萬 (762,803) | 1.71萬 (17,107) |
| 月流量變化量 | -22.52萬 (-22.8%) | -8233 (-32.5%) |
| 主要市場與市佔 | 🇺🇸 美國 (21.9%) | 🇺🇸 美國 (69.4%) |
| 每月直接造訪 | 37萬 | 7493 |
| 使用者評分 | 4.4 | 4.4 |
| 支援平台 | 網站 | 網站 |

前沿 AI 訓練所用的環境,以及負責例行工作的代理
涵蓋視覺、語言與生成任務的結構化標註、資料策展、模型評估與人機協作審閱,是訓練與評估資料的基礎。
Recursion 代理會從建置、告警、提交、檢查點與排程自動啟動,並在背景持續工作直到任務完成可審查。
專業代理涵蓋軟體工程、可靠性、資安、ML 研究、評估、研發、產品與財務,從修復不穩定測試到撰寫事故檢討草稿。
每次執行都由評分代理打分,評分後的工作階段留下記憶、筆記與技能,供下一次執行前讀取。

Unlearn.AI builds AI-generated digital twins of clinical trial participants — forecasts of how each individual patient's outcomes would evolve — and uses them to sharpen design, interim, and readout decisions in drug development.
Model-based forecasts of each participant's clinical outcomes, generated from baseline data already collected, that act as a virtual comparator or as additional information per enrolled patient.
A twin-based covariate adjustment approach that Unlearn describes as EMA-qualified and FDA-supported, used to reduce sample size in randomized controlled trials while maintaining power.
Build and compare design scenarios — endpoints, eligibility criteria, sample size, assumptions — so governance decisions are made with quantified tradeoffs instead of positions.
Pre-specified twin-based analyses added to the statistical analysis plan before unblinding, plus twin-adjusted analysis delivered at interim and at readout.
專項更強簡化從標註到模型部署的端到端電腦視覺工作流程,但缺乏 Labelbox 的一般企業代理編排功能。
提供強大的多模態標註與工作流程自動化,並為電腦視覺團隊提供更直觀的資料策展管線。

專項更強在互聯環境中追蹤、評估並交付 AI 模型與代理。


專項更強Focuses strictly on LLM evaluation, testing, and observability rather than raw data annotation and training dataset curation.
專項更強Focuses strictly on LLM evaluation, testing, and observability rather than raw data annotation and training dataset curation.
用 AI 找出符合篩檢條件的病患,並為每顆肺結節評風險分