~95% on SimpleQA (e.g. Qwen3.6-27B on a 3090). Supports all local and cloud LLMs (llama.cpp, Ollama, Google, ...). 10+ search engines - arXiv, PubMed, your private documents. Everything Local & Encrypted.
Python academiaanthropicarxivbravedeep-researchencryptionhome-automationhomeserverlocallocal-deep-researchlocal-llmmistralollamaopenaipubmedresearchresearch-toolretrieval-augmented-generationsearxngself-hosted
Text, CSV and TSV uploads that are not UTF-8 are read again instead of being refused: Windows-1252 ("ANSI") files, as Notepad and Excel write them on Western-European Windows, and UTF-16 or UTF-32 files with a byte-order mark. A UTF-8 byte-order mark no longer ends up in the text or in the first CSV column name. Files in other legacy encodings (Cyrillic, Chinese, Japanese, Korean) are still either refused or read as Windows-1252; the encoding a file was read in is recorded on each of its documents as detected_encoding.