@AmarSVS what are you using for model/reasoning/max depth etc when testing halo? I’ve done some runs for traces containing ~5k+ spans, doable on default settings but with low reasoning on gpt 5.5, not sure if I’m leaving anything on the table though
@adithya_s_k harbor ftw, it’s also great for local inference workloads too (flag for setting the api base url) which I’ve been using for eval’ing agents running over os weights you’re hosting yourself
A few weeks and 550 million gpt-5.4/5.5 tokens later... I ported the entirety of VS Code/Cursor's git diff view to Lua
Introducing glance - a standalone TUI for reviewing and actioning on git diffs
Glance supports:
- a side-by-side diff view
- stage, unstage, discard or commit staged changes from the filetree
- a config layer to customize everything from glance's keymaps to its theme
- a 3-way merge view for conflict resolution
...and more 👇
Glance was built across a lot of Codex sessions, with Claude Code running review passes. Every diff was also reviewed with... well, glance itself.
I honestly would not have been able to enumerate across countless git edge cases without AI, and even then git is gnarly and I'm sure I missed some (let me know!)