Atoqu Introduces Open-Source Search Engine Core
Atoqu, a new open-source search engine core written in pure C++17 with zero external dependencies, is preparing its v1.2 release and solicits community input on license choice and performance expectations. It implements GPU-accelerated vector search (CUDA and OpenCL), multi-mode ranking (Literal, Vector, Hybrid, BM25, Recency, TagBoost), embedding providers, and forward compatibility. The project requests feedback on licensing options (Apache 2.0, MPL 2.0, GPLv3, AGPLv3) and feature priorities for v2.0.
Key Points
- 1Implements GPU-accelerated vector search, multi-mode ranking, and pure C++17 zero-dependency core
- 2Offers potential higher performance and lighter deployment without JVM, Python, Lucene, or FAISS dependencies
- 3Invites community input on licensing, benchmarks, and v2.0 feature priorities like distributed indexing and crawler
Scoring Rationale
Relevant new open-source search core with GPU vectors, but limited by pre-release status and lacking independent benchmarks.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
