Singapore – – September 30, 2026 -- Querit has launched a dedicated Code Search capability for its Querit Search API, letting developers switch from general web search to programming-focused retrieval by setting a single parameter, Vertical=Code.
Internal tests show 81% of returned results are directly adoptable by coding agents
In internal, programming-focused evaluations covering tasks such as function generation, autocomplete, and debugging, 81% of results returned by the Querit Search API were directly adoptable by coding agents, as scored by an LLM-as-a-judge under identical token budgets for intent and constraint satisfaction.
Querit's base Search API already ranks #1 on the FreshQA benchmark with 83.17% accuracy
The underlying Querit Search API runs on a proprietary web index spanning hundreds of billions of pages with multilingual and cross-regional coverage, drawing on regulatory bodies, standards organizations, technical documentation, and professional communities. On FreshQA, a public benchmark for time-sensitive retrieval, it ranked #1 with 83.17% accuracy.
Three-axis redesign targets framework, runtime, and error-specific constraints
Querit rebuilt its coding vertical around three mechanisms: prioritizing official documentation, API references, SDK manuals, and communities such as Stack Overflow while downweighting stale secondary content; constraint-aware retrieval that identifies programming languages, framework versions, runtime environments, and error signatures to filter out semantically relevant but incompatible results; and adaptive, structured output that scales from single API how-tos to end-to-end tasks while preserving code blocks and parameters within a strict token budget.
Shawn Xu says vertical retrieval determines agent output quality
"General search decides what an agent sees; deep, vertical retrieval decides how good its output is," said Shawn Xu, Head of Global Product at Querit. "We're building a real-time information layer that truly understands task intent — and coding is where that architecture proves itself."
Existing Search API users can enable Code Search without reworking integrations
The Querit Search API and its code vertical are generally available worldwide as of today. Existing users can activate Code Search by setting the code vertical parameter in their request, without changing endpoints. Querit supports the Model Context Protocol and is integrated with agent frameworks including LangChain, Dify, RAGFlow, CAMEL-AI, Cherry Studio, Eigent, EigenFlux, Continua, ModelOS, and Atlas Cloud.
Coding coverage will next expand to major open-source repositories
Together with Querit's Contents API and Monitors API, the Search API forms a real-time information layer spanning web search, content extraction, and continuous monitoring. Querit plans to extend coding coverage to major open-source repositories next.