Search

More Filters

Show More
Add Listing
Go Back

blast

0
New
Multi-Agent Platforms & Frameworks
Report

Report Abuse

Basic Information

BLAST is a high-performance serving engine designed to add web browsing capabilities to AI applications. It exposes an OpenAI-compatible API so existing clients can send requests and receive streaming, browser-augmented LLM outputs. The project focuses on efficient resource management for interactive web-based agents by providing built-in concurrency, automatic parallelism, and caching to reduce latency and costs. BLAST can be run locally or deployed to serve multiple users while keeping memory and budget constraints under control. Typical use cases described include embedding web browsing AI into apps, automating web-based workflows, and local development and experimentation. The repository provides a pip-installable package and a serve command to run a local server and example code showing how to stream browser actions through a compatible OpenAI client.

Links

Stars
548
App URL
https://github.com/stanford-mast/blast
Github Repository
https://github.com/stanford-mast/blast/blob/main/README.md

App Details

Features
BLAST offers an OpenAI-compatible API as a drop-in replacement for OpenAI clients. It implements automatic parallelism and prefix caching to improve throughput and lower model usage costs. Streaming support lets clients receive real-time browser-augmented LLM outputs and browser action events. Built-in concurrency and efficient resource management enable serving many users simultaneously. The engine automatically caches and parallelizes requests to keep interactive latencies low. The README highlights a quick start workflow using a pip install and a serve command and includes a client example that streams response deltas from a base_url, demonstrating integration with existing OpenAI-style clients.
Use Cases
BLAST simplifies integrating web browsing into AI systems by handling low-level orchestration such as parallel browser actions, caching, streaming, and concurrency so developers can focus on application logic. It allows reuse of existing OpenAI client code by exposing a compatible API endpoint, reducing migration effort. Automatic caching and parallelization lower API and compute costs while maintaining interactive performance. Concurrency controls and resource management support scaling to multiple users without excessive memory or budget usage. The project supports local use for experimentation and development, includes documentation and contribution guidance, and is released under an MIT license to enable adoption and extension.

Categories

Content Creation (5042)
Monetization & Business (2246)
Development Tools (1407)
Marketing & Social Media (1403)
Education & Career (1380)
Multi-Agent Platforms & Frameworks (1009)
Health & Life (957)
AI Resources (920)
Agents & Automation (778)
Data & Analytics (764)
Productivity Tools (679)
Writing & Editing (666)

Similar Listings

listing image

virtual lab

0 (0 review)
listing image

mnemo

0 (0 review)
listing image

EdgeChains

0 (0 review)
listing image

RAG Agents Accelerator

0 (0 review)
listing image

LLM Zero to Hundred

0 (0 review)

Featured Listings

Tags

Please fill the required fields*