HarnessEvolve decouples execution from evolution and aligns failed runs against reference trajectories (produced with ground-truth answers) to diagnose first-error steps, cluster recurring failure modes, then edit the harness—prompts, skills, tools, and execution logic—behind quality and performance gates. On CloudCoreNetwork-QA with Qwen3.6-27B, the full system hit 86.9% (57.8% without reference trajectories) and beat the strongest baseline by 21.6 points.

Key Takeaways

  • ✓Method: align failures to reference trajectories to fix long-horizon credit assignment.
  • ✓Edit surface: prompts, skills, tools, scripts, and execution logic behind leak/bloat and regression gates.
  • ✓Result: 86.9% on CloudCoreNetwork-QA with Qwen3.6-27B (57.8% without refs); +21.6pp vs strongest baseline.
⚡

Key Decision Metrics at a Glance

TL;DR Verdict
🏆 Benchmark RankRank
Result: 86.9% on CloudCoreNetwork-QA with Qwen3.6-27B (57.8% without refs); +21.6pp vs strongest baseline.
Full Leaderboard→
💰 Cost & PlansCompare
Qwen 2.5 Coder 32B · in: $0.14 / out: $0.42
Compare Plans↗
🛠️ Ecosystem & IDERules
HarnessEvolve · agent harness · Self evolution
Setup IDE Rules↗
🧭

Turn your technical choice into a development budget

Compare 40 dev plans & simulate token costs vs $20/mo subscriptions

⚔️ Compare Model→💰 Compare plans & limits↗
Action HubReady to adopt this in production?

Benchmark side-by-side against alternatives, or calculate monthly token cost vs subscription break-even.

⚔️ Launch VS Compare→💳 View Plans↗
Up NextLeCun and Balestriero turn the Platonic Representation Hypothesis into a theorem: a 109M text JEPA aligns with vision models better than 72B LLMs
Read Now→