Perplexity's "Search as Code" lets AI models write their own search pipelines instead of calling fixed APIs
Quick Answer
Perplexity's 'Search as Code' allows AI models to autonomously create search pipelines in Python, outperforming OpenAI and Anthropic on benchmarks while reducing token costs by up to 85%.
Quick Take
This innovative approach enables enhanced filtering and deduplication within a sandbox environment.
Key Points
- AI models can now write their own search routines in Python.
- The new system outperforms OpenAI and Anthropic on key benchmarks.
- Token costs are reduced by up to 85% with this architecture.
- Filtering and deduplication are handled inside a sandbox environment.
- This approach marks a shift from rigid search APIs.
Source Excerpt
Perplexity's new "Search as Code" architecture dumps rigid search APIs and lets AI models write their own search routines in Python. By letting the agent handle its own filtering and deduplication inside a sandbox, the system beats OpenAI and Anthropic on key benchmarks, while cutting token costs by up to 85 percent.
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from The Decoder
See more →
An AI model programmed nonstop for 19 days on a single MirrorCode task that cost $2,600 to run
Epoch AI's MirrorCode benchmark reveals Claude Opus 4.7 as the leader with a 56% solve rate, reconstructing a 16,000-line toolkit in 14 hours. Despite this, all models tested struggle with the most complex tasks, highlighting limitations in current AI capabilities. The single task consumed $2,600 over 19 days, raising questions about cost-effectiveness in AI development.
