Skip to main content
ToolPotion
AI Advisor
Submit a Tool (Free)

cklxx/laya-browser

AI ModelsAI Models & LLMsOpen source · Web

cklxx/laya-browser is a feature extraction model, fine-tuned as a browser-agent decision head. It utilizes a non-autoregressive "System 1" decision model to pick operations and target elements in a browser environment. This model is designed for tasks requiring rapid, single-pass decisions, offering a drop-in replacement for existing browser automation solutions and supporting various real-site multi-step tasks.

View Model
Share
View Model
Share

The cklxx/laya-browser model is a specialized feature extraction model, boasting 0.3 billion parameters, designed to function as a browser-agent decision head. It is a fine-tuned version of the `laya` model (convaiinnovations/laya), which operates as a non-autoregressive "System 1" decision model. Unlike traditional autoregressive models, `laya` uses a single bidirectional encoder pass to answer several typed questions (choice, score, noul) with calibrated probabilities, without generating text.

Read More

This particular repository fine-tunes `laya` specifically for browser automation tasks, serving as a drop-in replacement for `browser-use/jev-ultrafast`. It processes `/v1/systemone` requests, which mirror `laya`'s `predict(state, questions)` function. In every step, the model performs one forward pass, taking approximately 22 milliseconds on a full GPU clock, to select the appropriate operation (such as CLICK, TYPE_TEXT, SELECT, PRESS_ENTER, SCROLL, or DONE) and its corresponding target element on a webpage.

The model was trained and evaluated locally on an RTX 4070 Ti SUPER (16 GB) without relying on paid APIs. For text input and dropdown value selection, it integrates with a local Qwen3-8B-AWQ model. The current version, v19s, is available at the repository root, with earlier checkpoints accessible in the commit history. Updates are only released when a new version demonstrates clear improvements.

Evaluation results highlight its performance across various suites. Suite C, which involves 27 multi-step tasks on held-out real sites (including search, filters, sort, and open operations), shows v19s achieving approximately 20–26% success, a significant improvement from v17s's 2/54. Suite B, with 18 tasks on 18 held-out sites, maintains a perfect 54/54 score for both v17s and v19s. In Suite A, comprising 16 original real-site tasks, v19s scored 39/48. For webgym held-out synthetic forms (7 kinds × 10), v19s achieved 30/70, doubling v17s's 15/70. The model also demonstrates strong performance in held-out decisions on unseen sites (WebChain) for click operation and target top-1 accuracy.

Latency for a single decision is typically 22–27 ms on an RTX 4070 Ti SUPER at full clock. However, if the GPU drops to idle clocks between decisions (due to waiting for pages to load), subsequent decisions can take 75–270 ms. Users can lock the minimum SM clock to mitigate this. The model is integrated using `transformers` and `torch`, shipping its own code. It can also be loaded via the upstream `laya` package. Key changes since v17s include harness improvements, such as waiting for navigation to complete, an "undo guard" to prevent repetitive actions, better handling of form field data, and scrolling options into view. Training data was also expanded with WebChain human trajectories and DAgger on webgym. Current limitations include stopping too early in multi-step tasks and issues with pages requiring extensive scrolling or sites blocking headless Chromium.

  • CategoryAI Models & LLMs
  • Pricing ModelOpen source
  • DeploymentCloud
  • PlatformsWeb

  • Non-autoregressive decision model for browser automation

  • Single bidirectional encoder pass for typed questions

  • Calibrated probabilities for choice, score, and noul answers

  • Drop-in replacement for browser-use/jev-ultrafast

  • Picks browser operations (CLICK, TYPE_TEXT, SELECT, PRESS_ENTER, SCROLL, DONE)

  • Identifies target elements for operations

  • Fast decision latency of ~22 ms on full GPU clock

  • Trained and evaluated locally on RTX 4070 Ti SUPER

  • Integrates with local Qwen3-8B-AWQ for text input and dropdown values

  • Supports multi-step tasks on real-world websites

  • Includes an "undo guard" to prevent repetitive actions

  • Improved handling of form field placeholders, types, and patterns

  • Offers options scrolled out of view by scrolling them into view

  • Trained with WebChain human trajectories and DAgger on webgym

  • Apache-2.0 license for the model code

  1. Access Model: Download the cklxx/laya-browser model from Hugging Face.

  2. Set Up Environment: Ensure `torch` and `transformers` libraries are installed in your Python environment.

  3. Integrate via API: Use the `decide` function to build requests in the model's trained format and run a forward pass.

  4. Utilize as Replacement: Implement `m.systemone(body)` to answer `jev-ultrafast /v1/systemone` requests.

  5. Load with Laya Package: Alternatively, load the model using `laya.load("cklxx/laya-browser")` with the upstream `laya` package.

  6. Apply Harness Improvements: Apply the `code/jev-ultrafast.patch` to `jev-ultrafast` for improved evaluation and training setup.

  7. Optimize Latency: Optionally, lock the GPU's minimum SM clock using `sudo nvidia-smi -lgc 2100,3135` to reduce decision latency.

  • Automated Browser Navigation
  • Web Data Extraction
  • Form Filling Automation
  • Website Testing
  • Agent-Based Web Interaction
  • Research Data Collection
  • Repetitive Online Tasks

cklxx/laya-browser Reviews

Loading...

Popular AI Tools Like cklxx/laya-browser

AutoTrust-JEV-27B v1.1 is a text decision model designed to assign probabilities to answer options for various questions. It excels in tool selection, intent classification, and…

AI Models & LLMs
View Tool

Text decision model designed to assign probabilities to answer options for various questions. It excels in tool selection, intent classification, and…AutoTrust-JEV-27B v1.1 is a text decision model designed to assign probabilities to answer options for various questions. It excels in tool selection, intent classification, and…

AI Models & LLMs
See the Top 5 AI Models

Clef

AI Models

Clef is a 27B multimodal model designed to transform a state and schema of typed questions into decisions. It processes inputs like text, JSON, images, or video, and returns…

AI Models & LLMs
View Tool

AI Models

27B multimodal model designed to transform a state and schema of typed questions into decisions. It processes inputs like text, JSON, images, or video, and returns…Clef is a 27B multimodal model designed to transform a state and schema of typed questions into decisions. It processes inputs like text, JSON, images, or video, and returns…

AI Models & LLMs
See the Top 5 AI Models

GLiDE 27B is a 27 billion parameter decision model developed by Fastino. It processes states containing text, JSON, and optional images to answer typed questions with…

AI Models & LLMs

Ranked #76 in LLMs

View Tool

27 billion parameter decision model developed by Fastino. It processes states containing text, JSON, and optional images to answer typed questions with…GLiDE 27B is a 27 billion parameter decision model developed by Fastino. It processes states containing text, JSON, and optional images to answer typed questions with…

AI Models & LLMs
Ranked #76 in LLMs

Jev Ultrafast Agent is a free browser extension for Chrome and Edge that automates tasks on web pages using your chosen AI model. It interacts with elements directly, rather than…

Browser & Web Automation
View Tool

Free browser extension for Chrome and Edge that automates tasks on web pages using your chosen AI model. It interacts with elements directly, rather than…Jev Ultrafast Agent is a free browser extension for Chrome and Edge that automates tasks on web pages using your chosen AI model. It interacts with elements directly, rather than…

Browser & Web Automation
See the Top 250 AI Apps

StartLux-Decision

AI GitHub Repos

StartLux-Decision offers open decision models ranging from 0.8B to 27B parameters, providing calibrated probabilities for each option. It generates concise thoughts when…

AI Models & LLMs
View Tool

AI GitHub Repos

Offers open decision models ranging from 0.8B to 27B parameters, providing calibrated probabilities for each option. It generates concise thoughts when…StartLux-Decision offers open decision models ranging from 0.8B to 27B parameters, providing calibrated probabilities for each option. It generates concise thoughts when…

AI Models & LLMs
See the Top 250 AI GitHub Repos

DeepSeek-v3 offers instant AI solutions powered by advanced MoE architecture and state-of-the-art language models. Experience cutting-edge AI capabilities with this stable, free,…

AI Models & LLMs
View Tool

Offers instant AI solutions powered by advanced MoE architecture and state-of-the-art language models. Experience cutting-edge AI capabilities with this stable, free,…DeepSeek-v3 offers instant AI solutions powered by advanced MoE architecture and state-of-the-art language models. Experience cutting-edge AI capabilities with this stable, free,…

AI Models & LLMs
See the Top 250 AI Apps

Browser Harness

AI GitHub Repos

Browser Harness is a self-healing framework that enables large language models (LLMs) to complete any task within a browser environment. It provides robust automation capabilities…

Browser & Web Automation

Ranked #78 in Browser & Web Automation

View Tool

AI GitHub Repos

Self-healing framework that enables large language models (LLMs) to complete any task within a browser environment. It provides robust automation capabilities…Browser Harness is a self-healing framework that enables large language models (LLMs) to complete any task within a browser environment. It provides robust automation capabilities…

Browser & Web Automation
Ranked #78 in Browser & Web Automation

Browser Use is an open-source platform for browser automation, designed for AI agents. It provides tools for web interaction, including undetectable browsers and proxy…

Browser & Web Automation

Ranked #36 in AI Agents

View Tool

Open-source platform for browser automation, designed for AI agents. It provides tools for web interaction, including undetectable browsers and proxy…Browser Use is an open-source platform for browser automation, designed for AI agents. It provides tools for web interaction, including undetectable browsers and proxy…

FeaturedBrowser & Web Automation
Ranked #36 in AI Agents
1 of 8