LetTokenBurn
Sign in
Verified100%Harbor nativeCodingAgentic / Tool Use

SWE-bench Pro

SWE-bench Pro: 731个更难软件工程任务

PaperRepoDatasetLeaderboard

How to run

ltb · swe-bench-pro
$ npm i -g lettokenburn-cli
$ ltb bench run scale-ai/swe-bench-pro@2 --benchmark swe-bench-pro -a claude-code -m anthropic/claude-opus-4-8
$ ltb bench push

Harbor-native: you can also call Harbor directly.

harbor
$ harbor run -d scale-ai/swe-bench-pro@2 -a claude-code -m anthropic/claude-opus-4-8

Runs happen on your machine with your own subscription; only the score summary is pushed. CLI docs

Scores

0 sourced entries

No sourced scores recorded yet.

Every number links to where it was published. Low-confidence rows are muted and need per-cell verification.

Reproduced by the community

All runs
Loading community runs…

About

SWE-bench Pro (SWE-bench Pro) — 731个更难软件工程任务. Scoring: Resolved%. Difficulty: 极高. Imported from benchmark_research.md (2026-06).