LetTokenBurnSWE-TaskFlow: SWE-Bench Pro Interactive Task Flows This is the unified data release for “Coding-Agent Benchmarks Should Match Their Users’ Task Flows” by Igor Slinko, Yaroslav Golubev, and Sergey Titov.
$ npm i -g lettokenburn-cli$ ltb bench info swe-taskflow
External harness. No Harbor adapter has been generated for this benchmark yet. `ltb bench info` prints the known links.
Runs happen on your machine with your own subscription; only the score summary is pushed. CLI docs
No sourced scores recorded yet.
Every number links to where it was published. Low-confidence rows are muted and need per-cell verification.
SWE-TaskFlow: SWE-Bench Pro Interactive Task Flows
This is the unified data release for “Coding-Agent Benchmarks Should Match Their Users’ Task Flows” by Igor Slinko, Yaroslav Golubev, and Sergey Titov. Code: JetBrains-Research/SWE-TaskFlow-Generation-Public — dataset generation and QA validation tooling, the exact generation and agent prompts, and worked examples of how issues become task flows. Agent trajectories: JetBrains-Research/SWE-TaskFlow-SWE-Bench-Pro-Trajectories —…