Draco
PulseAugur coverage of Draco — every cluster mentioning Draco across labs, papers, and developer communities, ranked by signal.
-
PPIO Fusion model uses multiple AI agents to beat top models at 1/10th cost
PPIO has launched its Fusion model gateway, which enhances AI capabilities by orchestrating multiple specialized models to work collaboratively on complex tasks. This approach aims to improve intelligence and reduce cos…
-
Proteus model offers robust LiDAR compression with 70% truncation tolerance
Researchers have developed Proteus, a novel compression model specifically designed for LiDAR point clouds. This model employs a unique strategy of separating the significant bit-planes (SIG) from the insignificant ones…
-
Draco: A Fast, Self-Hostable Web Scraper Alternative to Firecrawl
Draco is a new, self-hostable web scraping tool built in Rust, designed as a lighter alternative to services like Firecrawl and Browserbase. It aims to efficiently extract clean Markdown and metadata from web pages, inc…
-
New method preserves clusters in lossy compression of scientific particle data
Researchers have developed a novel technique to preserve the integrity of clusters within scientific particle data during lossy compression. This method operates on data compressed by existing tools like SZ3, ZFP, Draco…
-
OpenSquilla integration beats GPT-5.6-sol on DRACO benchmark for cost and quality
A new evaluation of the DRACO benchmark's Brave Search group shows that GPT-5.6-sol has entered the rankings with an average score of 63.99 and a cost of $1.71 per task. However, an integrated solution called OpenSquill…
-
New agentic routing paradigm uses data flywheel for specialized LLMs
Researchers have introduced Harness-Native agentic routing, a new paradigm for managing large language model agents. This approach addresses the increasing specialization of AI models by enabling an execution harness to…
-
OpenSquilla integrates multiple Chinese AI models to top benchmarks, outperforming Fable 5 and GPT-5.5 on cost
OpenSquilla has released version 0.5.0 Preview 1 of its AI Agent project, introducing a multi-model integration and collaboration feature. This new approach organizes multiple Chinese AI models, including DeepSeek v4, G…
-
OpenRouter launches Fusion API to mimic Claude Fable 5 with model collaboration
OpenRouter has launched Fusion API, a composite model that uses multiple AI models to replicate the capabilities of Anthropic's Claude Fable 5. This comes after the US government imposed export controls on Fable 5, maki…
-
OpenRouter Fusion API faces criticism for cost and speed
OpenRouter has launched Fusion, a multi-model routing API designed to combine responses from several large language models into a single output. While marketed as a cost-effective alternative to single frontier models l…