Substrate Blog
JSON generation, and a new kind of program
Fast, accurate JSON generation with LLMs is the glue we need to do less prompt engineering, and more software engineering.
LangChain vs. Substrate: Cognitive overhead
We'll compare a simple example chaining two LLMs. With Substrate it takes 60% less code than LangChain, and runs faster.
Case Studies: Substack runs inference at scale on Substrate
Substack was our first beta customer. Over the last 8 months, they've migrated all of their inference workloads to Substrate. Today, they regularly run huge, complex workloads across all of the content on their platform.
Hello world, from Substrate
We're launching out of stealth – and announcing our $8M Series Seed led by Lightspeed.
Page 1 of 1