ae 28395df957 [fix] fix absolute and % token counts (#1931) ...

- For absolute, use non-cached input + output.
- For estimating what % of the model's context window is used, we need
to account for reasoning output tokens from prior turns being dropped
from the context window. We approximate this here by subtracting
reasoning output tokens from the total. This will be off for the current
turn and pending function calls. We can improve it later.

2025-08-07 08:13:36 +00:00

..

src

[fix] fix absolute and % token counts (#1931)

2025-08-07 08:13:36 +00:00

tests

fix: run apply_patch calls through the sandbox (#1705)

2025-07-30 16:45:08 -07:00

Cargo.toml

Introduce --oss flag to use gpt-oss models (#1848)

2025-08-05 11:31:11 -07:00