Fine-tuning & Custom LLMs
Summary
Bland will fine-tune a custom model for your enterprise using transcripts from succesful prior calls. Then Bland will host that LLM and provided dedicated infrastrucure to enable phone conversations with sub-second latency.
Bland will also enable you to connect to a custom LLM & will host that LLM to drive latency down further.
Get in touch here.
Background on fine-tuning
Traditionally, most AI phone agents use private models from companies like OpenAI and Anthropic. Those LLMs are large, and perform best at following instructions and delivering high quality outputs. The downside, however, is they are very slow. Additionally, because they’re general models, their personality, tone, and overall capabilities are limited.
Conversely, open source models generally perform worse at a broad range of tasks. However, by fine-tuning an open-source model with examples of a given task, you can significantly improve it’s performance at that task, even surpassing the capabilties of top-of-the-line models like GPT-4.
How do I fine-tune with Bland?
To inquire about fine-tuning connect with the Bland team here.
During the initial conversation you will discuss:
- The format of data we’ll require
- How long fine-tuning will take (typically one week)
- Pricing (typically under five figures)
How do I bring my own LLM?
The Bland team will advise on connection method, requirements for the connection, etc.
Typically takes under 24 hours to set up after kickoff.
Inquire here.