Argus: Self-Evolving Agentic Runtime Achieves 78% on SWE-Bench Without Model Retraining
imjustnewatai · x · 2026-08-07
Recursive self-improvement could begin before a model ever retrains. A newly released paper introduces Argus, a persistent agentic runtime that improves while keeping GPT-5.5 weights completely frozen.
Argus assigns distinct Manager, Planner, Engineer, and Reviewer roles. It only saves memories, skills, and failed routes as reusable assets after strict evidence checks. Across 731 software tasks, Argus scored about 78% compared to 59% for Direct Copilot, though it used 1.41x more tokens overall.
Crucially, as verified experience accumulated, mature execution waves used 21% fewer input tokens and 15% less active time per task than startup waves. This demonstrates that the runtime can evolve its control policy around the model, generating high-quality trajectories to eventually train successor models.
Related event: Argus Agent Runtime Enables Self-Evolution Without Retraining(2 posts)→
More from coding & agent
- Opencode Hits 8 Trillion Daily Tokens, Rivaling Codex and Claude — ycombinator · 2026-08-07
- Tencent's InsightEmb: Training Agentic Experience Retrieval Using Only Math Data — _reachsumit · 2026-08-07
- Why No Programming Language for LLMs Yet? Developer Calls for AI-First Design — jfischoff · 2026-08-07
- Whatomate: Open-Source Platform Integrating WhatsApp with AI Chatbots — tom_doerr · 2026-08-07
- Pyromind Launches Automated RL Platform as Continuous Learning Becomes Industry Consensus — 机器之心 · 2026-08-07
- Useful Hermes Prompt Tip: Make Agents Verify Code Changes — alexcovo_eth · 2026-08-07