From 134b874f979e09f691ddbeb5c2684de3f0780f89 Mon Sep 17 00:00:00 2001 From: Cherrytest Date: Tue, 10 Mar 2026 12:17:01 +0000 Subject: [PATCH] Add evaluation results on SWE-Bench Verified (#42) - Add evaluation results on SWE-Bench Verified (7584eabf4ba1460402b4b7e95103d51acf724d2f) - Rename .eval_results/swe_bench_verified.yaml A to .eval_results/swe_bench_verified.yaml (f8ed508a03f6c8aadef1d53e02b553aab0446773) - Update .eval_results/swe_bench_verified.yaml (a2e5c2dbbd5128e04af0d7523d29f799b4b3f19f) - Update .eval_results/swe_bench_verified.yaml (6543d1bffd005fe14f52d269a39281c524bd63ea) Co-authored-by: Niels Rogge (batch 1/1) --- .eval_results/swe_bench_verified.yaml | 9 +++++++++ 1 file changed, 9 insertions(+) create mode 100644 .eval_results/swe_bench_verified.yaml diff --git a/.eval_results/swe_bench_verified.yaml b/.eval_results/swe_bench_verified.yaml new file mode 100644 index 0000000..a7c48b1 --- /dev/null +++ b/.eval_results/swe_bench_verified.yaml @@ -0,0 +1,9 @@ +- dataset: + id: SWE-bench/SWE-bench_Verified + task_id: swe_bench_%_resolved + value: 75.80 + source: + url: https://www.swebench.com/ + name: SWE-Bench official evaluation + user: nielsr + notes: high reasoning \ No newline at end of file