RSS

Tyler Williams

Karma: 2

Hello! I am the sole founder of VANTA Research—a bootstrapped AI research project focused building safe and resilient AI models optimized for human-AI collaboration. All of our published work has been made open source thus far, and in roughly 2 months has garnered 60k+ downloads on Hugging Face and Ollama across original model families.

I have several goals with VANTA Research, but among the first major stops is building a large (400B+) open source foundational model from scratch. I love learning, asking hard questions, and a good mystery.

DMs are always open.

VANTA Re­search Rea­son­ing Eval­u­a­tion (VRRE): A New Eval­u­a­tion Frame­work for Real-World Rea­son­ing

Tyler Williams18 Sep 2025 23:51 UTC
1 point
0 comments3 min readEA link

Hal­lu­ci­na­tions May Be a Re­sult of Models Not Know­ing What They’re Ac­tu­ally Ca­pable Of

Tyler Williams16 Aug 2025 0:26 UTC
1 point
0 comments2 min readEA link

The Inevitable Emer­gence of Black-Mar­ket LLM Infrastructure

Tyler Williams8 Aug 2025 19:05 UTC
1 point
0 comments2 min readEA link

Cog­ni­tive Stress Test­ing Gem­ini 2.5 Pro: Em­piri­cal Find­ings from Re­cur­sive Prompt­ing

Tyler Williams23 Jul 2025 22:37 UTC
1 point
0 comments2 min readEA link

Syn­thetic An­thro­pol­ogy: Can AI Offer a New Lens on Hu­man Be­hav­ior?

Tyler Williams18 Jul 2025 2:22 UTC
1 point
0 comments2 min readEA link

How Prompt Re­cur­sion Un­der­mines Grok’s Se­man­tic Stability

Tyler Williams16 Jul 2025 16:49 UTC
1 point
0 comments1 min readEA link

How Deep­Seek Col­lapsed Un­der Re­cur­sive Load

Tyler Williams15 Jul 2025 17:02 UTC
2 points
0 comments1 min readEA link

Ad­ver­sar­ial Prompt­ing and Si­mu­lated Con­text Drift in Large Lan­guage Models

Tyler Williams11 Jul 2025 21:49 UTC
1 point
0 comments2 min readEA link