Vectara
@vectara
Followers
6K
Following
6K
Statuses
2K
Vectara's AI Assistant/Agent solution is enterprise-ready, focusing on accuracy, explainability of results, and secure access control.
Palo Alto, CA
Joined May 2022
🎉 3 Million Downloads and Counting! The momentum is unstoppable! Vectara’s Hughes Hallucination Evaluation Model (HHEM), the engine behind Vectara’s Factual Consistency Score, has officially hit 3 MILLION downloads on Hugging Face! This milestone is more than just a number—it’s a powerful signal that enterprises and developers are prioritizing accuracy, trust, and scalability in AI. As GenAI continues to evolve, reducing hallucinations isn’t optional—it’s essential. Thank you to the growing global community pushing the boundaries of what's possible with RAG AI Assistants and advancing safer, more reliable AI systems. We’re proud to stand alongside you in this mission. Check out the leaderboard today! #GenAI #ReliableAI #HallucinationDetection #Developer #RAGaaS #AIForEnterprise
0
6
13
RT @atokdase: @kimmonismus @vectara Gemini 2.0 Flash is amazing. I've been really surprised at how good it is. Anyone who thinks Google is…
0
1
0
RT @AdvaitOnline: Gemini 2.0 Flash (GA) and 2.0 Pro (exp) models have the lowest hallucination rate on @vectara hallucination bench. https:…
0
53
0
RT @bobgourley: @kimmonismus @vectara Great seeing @vectara here! Incredible team (like @awadallah, @ofermend and so many others).
0
1
0
RT @kimmonismus: As far as hallucinations are concerned, we have a new winner: Gemini 2.0 Flash has overtaken o3-mini in the lowest rate of…
0
90
0
RT @ofermend: @julianshalaby96 @ai_for_success @vectara Sonnet-3.5 (from @AnthropicAI ) is definitely there, just didn't make into the top…
0
1
0
RT @lukestanley: From @vectara Hallucination Leaderboard, Gemini-2.0-Flash-001 scored a low 0.7% hallucination rate, 0.1% lower than o3-min…
0
4
0
🚨 DeepSeek-R1 Hallucinates More—And Vectara’s Research Proves It 🚨 Exciting news! @vectara was recently featured in @semafor for our latest research on AI hallucinations. 📰 In the article, "DeepSeek hallucinates alarmingly more than other AI models," journalist Rachyl Jones highlights our findings on DeepSeek-R1’s significantly higher hallucination rate—showing that R1 hallucinates 4x more than its predecessor, DeepSeek-V3. Our research, led by Ofer Mendelevitch (@ofermend), Forrest Sheng Bao (@forrestbao), and Chenyu Xu, used Vectara’s HHEM model to benchmark hallucination rates across leading reasoning models. While closed models from OpenAI and Google performed best, DeepSeek-R1 struggled due to fine-tuning challenges. 🔗 Read the full Semafor article here: You can also check out the original blog featured here:
0
1
7
RT @ofermend: @lmarena_ai @GoogleDeepMind Gemini-2.0-Flash shows fantastic performance and is also now the #1 on @vectara hallucination lea…
0
6
0
RT @ofermend: @Alibaba_Qwen Congrats on the release of QWen2.5-Max. We just added it to @vectara 's Hallucination Leaderboard, with a 2.9%…
0
2
0
RT @aicampai: 📢 Starting Soon! 🚀 Join 1,000+ AI devs at the first-ever event: DeepSeek+, we dive into @deepseek_ai : Unleashed & Integrate…
0
2
0
RT @mmarshall: 🔹 RAG: The pragmatic choice → Hallucinations are still a problem. @awadallah @Vectara found DeepSeek hallucinates 14% vs. O…
0
2
0
RT @mmarshall: 🔹 Distillation → Small, task-specific models > bloated LLMs. @Sam_Witteveen: Reasoning models are thinking too much. Compan…
0
2
0
⚙️ AI Agents: Your new productivity powerhouse! AI Agents take AI Assistants to the next level by acting on behalf of users—executing tasks like scheduling meetings or automating routine actions. Learn how to build safe and effective AI Agents with the Executive RAG Cookbook. 📖 What you’ll discover: ✅ How AI Agents identify, prioritize, and execute actions ✅ Ways to safely deploy action-taking tools ✅ Tips for integrating Agents into your workflow 👉 Get your free copy today! #GenerativeAI #RAGaaS #EnterpriseAI #Vectara #Developers #AgenticRAG
0
0
2
RT @ofermend: @demishassabis Congrats on the new Gemini models The new Gemini-2.0-Flash-001 just took first place in the @vectara hallucina…
0
1
0
RT @7VoltCrayon: @arblauvelt @glean @vectara (spoiler: I work there) but we have an internal AI work-search tool built on our own platform…
0
1
0
RT @ElecArtTech: @vectara Vectara, un taux d'hallucination 4x plus élevé que DeepSeek-V3 soulève pas mal d'inquiétudes.
0
1
0
RT @kimmonismus: o3-mini-high has hallucination rate of 0.8 percent according to the HuggingFace Leaderboard; first AI to go under one perc…
0
170
0