FeaturedSemiconductor
Learn how we built Mockingbird V3 MM, a multimodal model for semiconductor AI agents that can produce grounded answers and work through multi-step engineering tasks.
Read more
Vectara’s Hallucination Leaderboard has been a critical tool for tracking and comparing the factuality of large language models. But as AI models evolve, so must the way we measure them. We're excited to introduce a new, more granular hallucination leaderboard.
Why Microsoft’s Copilot is to GenAI like Windows-95 is to operating systems.
A costly lesson in why AI outputs must be verifiable
HHEM-2.1-Open is the de facto open-source hallucination evaluation model for RAG, but how much better is the commercial version available in Vectara’s platform?
A deeper dive on why DeepSeek R1 hallucinates more than DeepSeek V3
DeepSeek has shaken the AI industry with its release of Deepseek-R1, but it turns out that Deepseek-R1 has a hallucination problem.
Vectara’s Hallucination Evaluation Model surpasses 2 million downloads as the fight against LLM hallucinations continues
The upgraded HHEM-2.1 outperforms both GPT-3.5-Turbo and GPT-4 for hallucination detection and is powering our updated HHEM leaderboard.
