Crimson AI NewsA CrimsonLingua Network service
EN ع
← Back to news
Research paper Hugging Face

Persistent Project Worlds Enable Autonomous Software Evolution, New EvoX Genesis Approach Shows

AI By Crimson AI Hugging Face Papers 13 August 2026 · 00:00 6 views
Share: X Telegram

A new research paper introduces EvoX Genesis, which organizes long-horizon software development around a persistent project rather than persistent agents, enabling multi-day compiler construction and numerical module reimplementation with low cost and high performance.

Persistent Project Worlds Enable Autonomous Software Evolution, New EvoX Genesis Approach Shows

Key points

Complex software systems are built over timescales that exceed the lifespan of any single coding agent. Most agentic systems preserve continuity through persistent sessions, memories, managers, or shared context. A new paper introduces EvoX Genesis (Genesis), which instead makes the software project persistent while keeping local agents finite-lived.

Genesis represents software as a persistent recursive world: each local world is situated by an accepted version and a repository path. Finite-lived agents propose local changes, recursive delegation moves work across paths, and only accepted consequences advance the persistent version history. This organization was evaluated across formation, continuation, and redevelopment scenarios.

Starting from a repository with no compiler implementation, Genesis used DeepSeek V4 Flash to build a Rust-based C compiler with about 250k tracked lines. The run lasted over 120 hours, archived over 1,000 agent episodes, and incurred only US$44 in model-token charges. The compiler passed the complete c-testsuite and most LLVM and Csmith tests.

In a separate compiler world generated with GLM 5.2, development continued after repeated agent replacement while retaining full test performance. Genesis also reimplemented 13 MESA modules with over 100k Fortran lines as a Rust workspace with nearly 90k Rust lines; across six numerical workloads, it achieved median speedups of 1.55–6.87x.

These results suggest that long-horizon software development can be organized around a persistent project rather than a persistent agent. One only needs to describe what the software should become; Genesis recursively creates local responsibilities, instantiates finite-lived agents where needed, validates what comes back, and carries accepted consequences forward as one persistent software world, scaling to millions of lines of code.

MetricValue
Compiler lines (Rust)~250k
Run duration120+ hours
Agent episodes archived1,000+
Model-token chargesUS$44
MESA modules reimplemented13
Fortran lines (original)100k+
Rust lines (new)~90k
Median speedups (6 workloads)1.55–6.87x
Source
Hugging Face · Hugging Face Papers
Related news
Research paper
Hugging Face 31 Aug 2026

Hugging Face Unveils StepGuard: Step-Level Guardrails for Safer AI Agents

StepGuard, a new step-level guard model from Hugging Face, audits agent actions before execution, reducing attack success rates by...

0
Research paper
Hugging Face 31 Aug 2026

Hugging Face Researchers Unveil ABot-Recon for Stable Long-Horizon 3D Reconstruction

ABot-Recon, a new streaming 3D reconstruction model from Hugging Face, achieves stable long-horizon performance using only local t...

0
Research paper
Hugging Face 31 Aug 2026

ContextPilot: Teaching Agents Proactive Context Management via Fine-Grained RL

Hugging Face researchers introduce ContextPilot, a framework that enhances long-horizon agent reasoning by expanding context-editi...

0