steve/llama-swap
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

proxy: compress captures with zstd (#668)

Browse Source 
The previous captures were saved uncompressed in memory. In agentic
workflows there can be many turns with each request containing the
previous context in the body with a lot of redundant data. Use zstd to
compress the request and response data before keeping a copy of memory.

Results: 

- Average Percentage Saved: 73.19%
- Average Compression Factor: ~6.77:1
This commit is contained in:
Benson Wong
2026-04-17 23:29:37 -07:00
committed by GitHub
parent c3f0d43e6e
commit 5e3c646829
5 changed files with 205 additions and 79 deletions
Download Patch File Download Diff File Expand all files Collapse all files
go.mod
+1
View File
@@ -6,6 +6,7 @@ require (
github.com/billziss-gh/golib v0.2.0
github.com/fsnotify/fsnotify v1.9.0
github.com/gin-gonic/gin v1.10.0
github.com/klauspost/compress v1.18.5
github.com/stretchr/testify v1.9.0
github.com/tidwall/gjson v1.18.0
github.com/tidwall/sjson v1.2.5