Microsoft Research Releases Webwright: A Terminal-Native Web Agent Framework That Scores 60.1% on Odysseys, Up from Base GPT-5.4’s 33.5%
Quick Answer
Microsoft Research has launched Webwright, a terminal-native web agent framework that utilizes reusable Playwright scripts.
Quick Take
Microsoft Research has launched Webwright, a terminal-native web agent framework that utilizes reusable Playwright scripts. This framework, powered by GPT-5.4, achieves a score of 60.1% on the Odysseys benchmark, significantly improving from the base model's 33.5%, and scores 86.7% on Online-Mind2Web, marking it as the top performer among open-source harness recipes.
Key Points
- Webwright replaces click-trace web automation with reusable Playwright scripts.
- Achieves 60.1% on Odysseys, up from GPT-5.4's 33.5%.
- Scores 86.7% on Online-Mind2Web, the highest among open-source recipes.
- Utilizes a single agent loop across three modules with about 1,000 lines of code.
- Targets developers seeking efficient web automation solutions.
Article Excerpt
From source RSS / original summaryMicrosoft Research introduces Webwright, a terminal-native browser agent framework that replaces click-trace web automation with reusable Playwright scripts. Using a single agent loop across three modules and roughly 1,000 lines of code, Webwright powered by GPT-5. 4 reaches 60. 1% on the long-horizon Odysseys benchmark and 86. 7% on Online-Mind2Web — the highest AutoEval score among open-sourced harness recipes. The post Microsoft Research Releases Webwright: A Terminal-Native Web Agent Framework That Scores 60.
1% on Odysseys, Up from Base GPT-5. 4’s 33. 5% appeared first on MarkTechPost.
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from MarkTechPost
See more →Meet Flash-KMeans: An IO-Aware, Exact K-Means That Runs Over 200× Faster Than FAISS on GPUs
Flash-KMeans is an open-source, IO-aware k-means implementation that operates over 200× faster than FAISS on NVIDIA H200 GPUs. It achieves 17.9× end-to-end and 33× speedup over cuML by optimizing distance calculations and updating mechanisms without approximating results. This advancement significantly enhances performance for data scientists and machine learning practitioners.