A new open-source system named LEVI has been developed to emulate AlphaEvolve's capabilities at a significantly reduced cost, reportedly up to 35 times cheaper. LEVI's core principle is that smaller language models can achieve comparable or superior results to larger ones through optimized search architectures and intelligent routing. The system has demonstrated strong performance in code and prompt optimization tasks, outperforming existing frameworks on benchmarks like ADRS and IFBench while using fewer computational resources. AI
IMPACT This system could enable more accessible and cost-effective AI development and experimentation by leveraging smaller models.
RANK_REASON The cluster describes a new open-source system that offers an alternative approach to existing AI optimization frameworks, focusing on cost-efficiency and performance with smaller models.
- AlphaEvolve
- Claude Code
- Claude Opus
- Codex
- LEVI
- Qwen-30B
- GPT-5
- HotpotQA
- IFBench
- Qwen3-30B-A3B
- TPU Research Cloud
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →