
ATLAS Trades Stocks by Rewriting Its Own AI Prompts as Weights
ATLAS wraps 25+ debating LLM agents in a Darwinian prompt-evolution loop, using Sharpe ratio as a loss function to self-improve trading strategies.
20 stories

ATLAS wraps 25+ debating LLM agents in a Darwinian prompt-evolution loop, using Sharpe ratio as a loss function to self-improve trading strategies.

vLLM-Omni brings production-grade serving to any-to-any multimodal models, cutting job completion time by up to 91% with a fully disaggregated pipeline.

A dual-path multimodal model that unifies text-to-image, image editing, and subject-driven generation with state-of-the-art open-source consistency.

The European Space Agency's Phi-lab releases a 24B parameter model, curated corpora, benchmarks, and a deployed RAG system for Earth Observation science.

A pure C and Metal engine streams a 397B parameter Qwen model from SSD on a 48GB MacBook, hitting 4.4 tokens per second with tool calling intact.

Sentrux gives AI coding agents a real-time architectural score so they can self-correct before your codebase becomes ungovernable

A TypeScript dashboard that plugs into an OpenClaw Gateway so you can chat with agents, approve tool calls, and schedule jobs from a browser.

Tesslate releases OmniCoder-9B in GGUF format, a Qwen3.5-based coding agent trained on 425K agentic trajectories from frontier models.

Prism ML's Bonsai-8B squeezes a full 8B language model into 1.15 GB using end-to-end 1-bit quantization, running 6x faster on RTX 4090 and reaching iPhone-class deployment.

Crucix is a self-hosted, open-source intelligence terminal that fuses 27 live data feeds — from satellite fire detection to conflict events — into a single Jarvis-style dashboard with LLM-powered alerts.

Pixelle-Video is a free, open-source pipeline that turns a single text topic into a fully edited short video — script, visuals, voice, and music included

Arnis is a free, open-source Rust tool that turns any OpenStreetMap location into a playable Minecraft world, complete with real terrain elevation, roads, and buildings

A prompt-only framework turns Claude Code and Codex into self-supervising agents that run for hundreds of hours without human babysitting

VideoRAG brings RAG to video at scale, with a desktop app that can query hundreds of hours of footage using a knowledge graph architecture

Baidu's new 4B parameter open-weight vision-language model tops OmniDocBench v1.5 and OCRBench while unifying parsing, layout, and understanding in one pass.

Alibaba open-sources the sandbox infrastructure it runs internally, giving AI agent builders a free, self-hostable alternative to managed platforms like E2B

Sure is a free, self-hostable personal finance app born from the ashes of Maybe Finance, now maintained by the open-source community

A 45-tip open-source playbook for Claude Code covers voice input, context management, container sandboxing, and multi-model orchestration — all battle-tested in production.

Stanford's NAV Lab took first place in NASA's Lunar Autonomy Challenge with a modular SLAM stack that hits centimeter-level accuracy in a GPS-denied simulated moonscape.

A multi-agent system with human checkpoints orchestrates causal analysis across relational databases, cutting data exploration time by 76 percent and dramatically improving effect estimation accuracy.