Researcher open-sources method to distill expensive LLMs into smaller models

Researcher open-sources method to distill expensive LLMs into smaller models

Hacker News·2h·babelfish

Babelfish published a technique for extracting knowledge from proprietary large language models—like OpenAI's—without access to their internals. By treating them as black boxes and using smart prompting, makers can now create smaller, cheaper models that replicate the larger ones' behavior, lowering the barrier to running capable AI locally.

Share𝕏Reddit

Related stories