Back to Arena

Nabla Copilot

by Nabla

System Card

OrganizationNabla
Released2023-03
Architectureagentic-workflow / Privacy-First LLM + STT Clinical Notes
DetailsFounded in Paris in 2018 by former Meta AI Research engineers, Nabla combines open-source and proprietary LLMs with custom STT models for clinical note generation in under 20 seconds. The platform's distinctive privacy stance: audio, transcript, and notes are only briefly stored in the clinician's browser — patient information never touches Nabla's servers. Integrates with major EHRs across 130+ health systems in 35+ languages. Launched the Copilot Chrome extension in March 2023 using GPT-3 for initial note drafting.
Parameters—
Domainepisodic-session
Open SourceNo
WebsiteVisit
healthcareprivacy-firstambient-scribeno-server-storagemulti-languagefrench

Capability Profile

Benchmark Scores

4 of 14 benchmarks
Long-Context Retrieval
0/5
RULER
no data
NIAH
no data
LooGLE
no data
LongBench
no data
∞Bench
no data
Multi-Turn Recall
2/2
LoCoMo
80.789p
MemoryBank
77.460p
Cross-Session Memory
1/1
LongMemEval
80.482p
Multi-Hop QA
0/3
BABILong
no data
MultiHop-RAG
no data
HotpotQA
no data
Agent Task Memory
1/1
AgentBench-Mem
7226p
Personalization
0/1
PerLTQA
no data
Factuality / Grounding
0/1
RAGAS
no data
Sources:Nabla Copilot vendor documentation; evaluated on AgentBench Memory Track (Tsinghua KEG, 2308)Nabla Copilot vendor documentation; evaluated on LoCoMo: Long-Term Conversational Memory Benchmark (Snap Research, 2402)Nabla Copilot vendor documentation; evaluated on LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory (Salesforce AI Research, 2410)Nabla Copilot vendor documentation; evaluated on MemoryBank: Enhancing LLMs with Long-Term Memory (Sun Yat-sen University, 2305)