← Front page

The a16z Show · Friday, July 31, 2026

Enterprises Face Hurdles in Adopting Fine-Tuned Open Source AI Models

While fine-tuning open-source models offers benefits, Ashwin Srinivasa notes that enterprises face challenges in adoption due to the non-trivial nature of the process. This includes acquiring specific data and developing custom evaluation benchmarks, which requires significant effort and specialized teams.

personAshwin Srinivasa

The tape

3 quotes
But it'll probably take longer than people think because even with our team, fine-tuning these models is non-trivial.
Ashwin Srinivasa
It's not just, oh, you can, so you're like, okay, we made the decision to use open source. Let's just use open source. Like, you have to get the data and then more importantly, you have to like have good evalues.
Ashwin Srinivasa
And if you think about our evalues, our evalues are very specific to us. You can't just like use some public eval set and like that just does the job.
Ashwin Srinivasa
Heard on The a16z Show — “How Enterprise AI Really Gets Deployed, published Friday, July 31, 2026. Heardvine summarizes and quotes with attribution and timestamps, and links to the original everywhere.
Transcribed via Gemini audio transcription · $0.07
Enterprises Face Hurdles in Adopting Fine-Tuned Open Source AI Models — Heardvine