AI Briefing#open-source
The latest AI and developer news about #open-source, with the original source and a short summary.
The latest AI and developer news about #open-source, with the original source and a short summary.
The open-source benchmark DungeonBench has been released to evaluate LLMs' navigation capabilities in simple dungeon games.
NeoMME, a multilingual multimodal encoder that processes text and images with a single Transformer without a vision tower, has been released.
llama.cpp now supports running the Nemotron-3-Puzzle-75B-A9B model, which adopts NVIDIA's hybrid MoE architecture.
MirroS released the Code-as-World model and research results, which restore videos into simulator-executable code to enhance physical reasoning performance.
zg, a hybrid local search infrastructure designed to enhance code and document search efficiency for AI agents, has been open-sourced.
Anthropic has released implementation code for shopping and merchant agents designed for retail and travel platforms.
Hugging Face has released 'funes', a local open-source memory tool that preserves cross-session reasoning for coding agents like Claude Code.
An open-source reproduction project has been released for a watercolor-generating coding model trained with human aesthetic preferences as the reward function, utilizing TRL and OpenEnv.
NanoCodana, a Claude Code-style AI coding agent that runs exclusively within browser tabs without a server, has been released as open source.
Google has released Mantis, an open-source framework that uses AI to automatically detect and patch vulnerabilities.
Jasper Research released an open-source cookbook covering the principles to implementation of text-to-image models, along with a 100 million image dataset.
An evaluation of six open-source AI detection models revealed that four failed to meet the 0.5% false positive rate threshold, with detection rates for humanized text peaking at 42%.
Amagine3D has released an open-source agent that generates editable parametric CAD sources from natural language and dimensions.
AWS Labs has open-sourced CAO, a multi-agent orchestrator that enables parallel or sequential control of various AI coding CLIs, including Claude Code and Codex.
TrueForge has released an open-source agent execution environment that integrates LLM tool calling, sandboxing, and session management.
aimake, an incremental build system that reduces unnecessary recomputation in AI/ML pipelines, has been released.
A model fine-tuned on image deraining tasks using the depth estimation backbone and neck weights of YOLO26 has been released.
Amazon has open-sourced 'Kiro Crew', an AI coding client previously used internally, under the Apache 2.0 license.
Vellium v1.1.0 released an update supporting local STT/TTS and automatic llama.cpp detection.
A small transformer trained for 1.5 hours achieved 44% performance on the ARC-AGI-1 benchmark at a cost of 67 cents.
Rakazo, an open-source AI agent platform that allows users to directly control models and execution environments, has been released in beta.
Hugging Face has released 207 optimized WebGPU kernels and a loader library to enhance in-browser AI inference performance.
Google combined Antigravity's Teamwork framework with Gemini 3.7 Flash to solve complex multi-agent problems.
MIT researchers released 'research-graph', an open-source CLI tool that validates the integrity of multi-agent workflows.