Skip to content

v0.5.2

Choose a tag to compare

View all tags
@chopratejas chopratejas released this 20 Mar 19:56
· 1515 commits to main since this release
v0.5.2
b5631cd

What's Changed

  • Fix: Cache-aware cost pricing — Dashboard "Input cost" now uses LiteLLM's native cache pricing (cache reads at 10%, cache writes at 125% for Anthropic) instead of list price for all tokens. Previously overstated input costs significantly.
  • Dashboard clarity — Added note that cost estimates cover message tokens only (excludes system prompt & tool definitions).
Assets 2
Loading