TerminalBenchGenerator: logprobs + session ID#448
Merged
tyler-griggs merged 5 commits intoNovaSky-AI:mainfrom Oct 14, 2025
Merged
TerminalBenchGenerator: logprobs + session ID#448tyler-griggs merged 5 commits intoNovaSky-AI:mainfrom
tyler-griggs merged 5 commits intoNovaSky-AI:mainfrom
Conversation
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Code Review
This pull request adds support for logprobs and session_id in TerminalBenchGenerator. The changes look good overall, but I have identified a few areas for improvement. Specifically, I've suggested clarifying an error message, adding a validation check to prevent data misalignment between tokens and logprobs, and fixing a potential bug related to how logprobs are handled when they are empty. These changes will improve the robustness and debuggability of the new functionality.
skyrl-train/examples/terminal_bench/generator/terminal_bench_generator.py
Show resolved
Hide resolved
skyrl-train/examples/terminal_bench/generator/terminal_bench_generator.py
Outdated
Show resolved
Hide resolved
skyrl-train/examples/terminal_bench/generator/terminal_bench_generator.py
Outdated
Show resolved
Hide resolved
erictang000
added a commit
that referenced
this pull request
Oct 14, 2025
Fixes ci failure after #448
li-boxuan
pushed a commit
to li-boxuan/SkyRL
that referenced
this pull request
Nov 23, 2025
Fixes ci failure after NovaSky-AI#448
dzorlu
pushed a commit
to fleet-ai/SkyRL
that referenced
this pull request
Feb 4, 2026
With laude-institute/harbor#45, Sandboxes now returns logprobs for terminus agent, so TerminalBenchGenerator could leverage it if applicable. This PR doesn't enable `trainer.algorithm.use_tis=true` so it should be a no-op. With laude-institute/harbor#50, Terminus agent now accepts "session_id" parameter, and it will show up in the litellm request body. TerminalBenchGenerator could leverage this for better routing. Note that I don't have access to GPU yet so this is not tested. --------- Co-authored-by: Tyler Griggs <131809874+tyler-griggs@users.noreply.github.com>
dzorlu
pushed a commit
to fleet-ai/SkyRL
that referenced
this pull request
Feb 4, 2026
Fixes ci failure after NovaSky-AI#448
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
With laude-institute/harbor#45, Sandboxes now returns logprobs for terminus agent, so TerminalBenchGenerator could leverage it if applicable. This PR doesn't enable
trainer.algorithm.use_tis=trueso it should be a no-op.With laude-institute/harbor#50, Terminus agent now accepts "session_id" parameter, and it will show up in the litellm request body. TerminalBenchGenerator could leverage this for better routing.
Note that I don't have access to GPU yet so this is not tested.