Skip to main content
Startups 2 min read

Yupp Shuts Down After Raising $33 Million From a16z Crypto

Yupp, the crowdsourced AI model-picking startup backed by a16z crypto's Chris Dixon, shuts down less than a year after launch despite 1.3 million users.

Yupp AI startup shuts down after raising 33 million dollars
Yupp AI startup shuts down after raising 33 million dollars
  • Pankaj Gupta and Gilad Mishne announce the closure of Yupp, their crowdsourced AI model-comparison startup.
  • The company raised a $33 million seed round in 2024 led by a16z crypto’s Chris Dixon, with 45+ angel investors including Jeff Dean and Aravind Srinivas.
  • Yupp signed up 1.3 million users and collected millions of AI model preferences monthly but failed to reach product-market fit.
  • The shutdown highlights how fast-moving AI model improvements can kill a startup’s value proposition overnight.

1.3 Million Users Weren’t Enough

Yupp let consumers test and compare results from 800 AI models for free, including state-of-the-art offerings from OpenAI, Google, and Anthropic. Users would receive multiple replies to a single prompt and vote on which model performed best. The idea was to generate anonymized preference data that model makers would pay for. The startup said it signed up 1.3 million users, collected millions of preferences monthly, and even had a few AI labs as paying customers.

It wasn’t enough. Co-founders Gupta and Mishne announced the shutdown on Tuesday, less than a year after launching. The $33 million seed round led by a16z crypto’s Chris Dixon had been a giant raise for its time. The cap table was stacked: Google DeepMind chief scientist Jeff Dean, Twitter co-founder Biz Stone, Pinterest co-founder Evan Sharp, and Perplexity CEO Aravind Srinivas all wrote checks.

AI Models Moved Faster Than the Business

The core problem was speed. AI models improved by such leaps and bounds over the past few months that the gap between them narrowed, eroding the need for a consumer-facing comparison tool. “The AI model capability landscape has changed dramatically in the last year alone,” Gupta wrote on X. “The future is not just models but agentic systems.”

The market for AI feedback also shifted underneath Yupp. Companies like Scale AI and Mercor pioneered a model where specialty experts — PhDs and domain specialists — are hired directly into the reinforcement learning loop. Labs are paying for expert-grade signal, not crowdsourced consumer votes. And as Silicon Valley pivots toward agentic AI built for other AIs, consumer feedback on model quality becomes even less relevant. Some of Yupp’s employees are joining a “well-known” AI company. The rest are looking for their next role.

TechCrunch | Pankaj Gupta on X | Chris Dixon on X

Tags

#Yupp #a16z #AI #Shutdown #Startups

More in Startups