Back to Arena

Saner AI

by Saner.AI

System Card

OrganizationSaner.AI
Released2023-11
Architecturevector-rag / AI Second Brain with Semantic Capture
DetailsSaner.AI embeds as a persistent browser side panel that automatically highlights, saves, summarizes, and tags web content as you browse. Notes are stored in a personal knowledge base with semantic search, AI tag suggestion, and similar-note surfacing via vector embeddings. A personal AI chat layer lets users synthesize knowledge across saved notes, GPT-4, and Gemini with web search augmentation.
Parameters—
Domainrag-retrievalpersonalization
Open SourceNo
WebsiteVisit
second-brainbrowser-extensionadhdsemantic-capturenote-takingpersonal-knowledge

Capability Profile

Benchmark Scores

6 of 14 benchmarks
Long-Context Retrieval
1/5
RULER
no data
NIAH
no data
LooGLE
no data
LongBench
603p
∞Bench
no data
Multi-Turn Recall
2/2
LoCoMo
57.78p
MemoryBank
65.717p
Cross-Session Memory
1/1
LongMemEval
59.19p
Multi-Hop QA
2/3
BABILong
no data
MultiHop-RAG
64.621p
HotpotQA
60.413p
Agent Task Memory
0/1
AgentBench-Mem
no data
Personalization
0/1
PerLTQA
no data
Factuality / Grounding
0/1
RAGAS
no data
Sources:Saner AI vendor documentation; evaluated on HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering (Stanford / CMU, 1809)Saner AI vendor documentation; evaluated on LoCoMo: Long-Term Conversational Memory Benchmark (Snap Research, 2402)Saner AI vendor documentation; evaluated on LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding (Tsinghua KEG, 2308)Saner AI vendor documentation; evaluated on LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory (Salesforce AI Research, 2410)Saner AI vendor documentation; evaluated on MemoryBank: Enhancing LLMs with Long-Term Memory (Sun Yat-sen University, 2305)Saner AI vendor documentation; evaluated on MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries (HKUST, 2401)