{"slug": "ai-agents-weekly-nvidia-avo-trueforge-chroma-foundation-fragile-self-improvement", "title": "🤖 AI Agents Weekly: NVIDIA AVO, TrueForge, Chroma Foundation, Fragile Self-Improvement, Ornith-1.5, dots3-note, DeepSeek Vision, and More", "summary": "NVIDIA's general-purpose coding agent AVO scored a perfect 100.00 RHAE on the ARC-AGI-3 public set, clearing all 183 levels across 25 environments with no instructions, rules, or stated goals, while Claude Opus 5 alone scores roughly 30% on the same benchmark. TrueFoundry open-sourced TrueForge under an MIT license, a vendor-neutral agent harness that reports roughly 30% lower cost on identical tasks with the same model on DevRev's Enterprise-Bench, and up to 75% savings when routing to open-source models with accuracy matched.", "body_md": "In today’s issue:\n\nNVIDIA AVO hits 100% on ARC-AGI-3\n\nTrueFoundry open-sources TrueForge harness\n\nChroma ships Foundation agent memory\n\nSelf-improvement gains vanish when reshuffled\n\nOrnith-1.5 self-improves to Opus level\n\ndots3-note runs for days\n\nDeepSeek adds vision to V4-Flash\n\nStripe acquires OpenRouter\n\nCursor rebuilds Git storage as a database\n\nHarvey post-trains Tenet on Kimi K3\n\nSlack Code makes coding multiplayer\n\nLEGO-RL trains agents in Claude Code\n\nAnd all the top AI dev news, papers, and tools.\n\n## Top Stories\n\n### NVIDIA AVO Solves ARC-AGI-3\n\nNVIDIA’s general-purpose coding agent AVO scored a perfect 100.00 RHAE on the ARC-AGI-3 public set, clearing all 183 levels across all 25 environments with no instructions, rules, or stated goals.\n\n**The harness carries the result:** Claude Opus 5 alone scores roughly 30% on the same benchmark. Wrapping it in AVO takes it to 100%, which NVIDIA frames as evidence that system design, not model capability alone, unlocks frontier long-horizon performance.**Agentic variation loop:** AVO cycles through inspect context, plan the next change, implement, evaluate with a scoring function, then diagnose and repair from failed attempts, committing accepted candidates into a growing solution lineage.**Persistent memory and a supervisor:** Prior implementations, evaluation results, compiler output, and reasoning are retained so the agent resumes from current state, while a separate monitor watches for stagnation and conditionally redirects the main agent.**Transfer across domains:** The same architecture was originally built for CUDA GPU kernel optimization and moved to interactive reasoning unchanged, using about 12% fewer environment actions than the prior VISTA baseline.\n\n### TrueFoundry Open-Sources TrueForge\n\nTrueFoundry open-sourced TrueForge under an MIT license, a vendor-neutral agent harness that acts as the runtime layer turning an LLM into a working agent.\n\n**Batteries included:** MCP tools, a skills registry, sandboxed execution, human-in-the-loop approval gates for sensitive actions, subagents, durable state for long-running tasks, and step-level tracing all ship in the box.**Vendor-neutral by design:** It runs on your own infrastructure against any commercial or open-source model, with per-task switching, and the TrueFoundry AI Gateway is optional rather than required.**Cost is the pitch:** On DevRev’s Enterprise-Bench, evaluated blind, TrueFoundry reports roughly 30% lower cost on identical tasks with the same model, and up to 75% savings when routing to open-source models with accuracy matched.**Traction:** The repo has cleared 2,800 stars since the August 19 launch. Star it if you want to follow the project.", "url": "https://wpnews.pro/news/ai-agents-weekly-nvidia-avo-trueforge-chroma-foundation-fragile-self-improvement", "canonical_source": "https://nlp.elvissaravia.com/p/ai-agents-weekly-nvidia-avo-trueforge", "published_at": "2026-08-22 15:22:20+00:00", "updated_at": "2026-08-22 15:43:11.056492+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-agents", "ai-research", "ai-tools"], "entities": ["NVIDIA", "ARC-AGI-3", "Claude Opus 5", "TrueFoundry", "TrueForge", "DevRev", "Enterprise-Bench"], "alternates": {"html": "https://wpnews.pro/news/ai-agents-weekly-nvidia-avo-trueforge-chroma-foundation-fragile-self-improvement", "markdown": "https://wpnews.pro/news/ai-agents-weekly-nvidia-avo-trueforge-chroma-foundation-fragile-self-improvement.md", "text": "https://wpnews.pro/news/ai-agents-weekly-nvidia-avo-trueforge-chroma-foundation-fragile-self-improvement.txt", "jsonld": "https://wpnews.pro/news/ai-agents-weekly-nvidia-avo-trueforge-chroma-foundation-fragile-self-improvement.jsonld"}}