Auto-update results.csv

results.csv

@@ -5,10 +5,13 @@ SWE-Bench Pro -- Ansible,ScaleAI/SWE-bench_Pro,96,https://huggingface.co/dataset
 SWE-Bench Verified,SWE-bench/SWE-bench_Verified,500,https://huggingface.co/datasets/SWE-bench/SWE-bench_Verified,Claude Code,[],False,https://github.com/anthropics/claude-code,Gemma4-31B-FP8,RedHatAI/gemma-4-31B-it-FP8-block,True,31,fp8,https://huggingface.co/RedHatAI/gemma-4-31B-it-FP8-block,harbor,https://github.com/harbor-framework/harbor,500.0,35.0,0.612,651306166.0,0.0,4591859.0,655898025.0,524857.0,669355.0,104.14,1302612.0,0.0,9183.0,1311796.0,0.21,1338.0,1049.0
 SWE-Bench Verified,SWE-bench/SWE-bench_Verified,500,https://huggingface.co/datasets/SWE-bench/SWE-bench_Verified,OpenCode,[],True,https://github.com/anomalyco/opencode,Gemma4-31B-FP8,RedHatAI/gemma-4-31B-it-FP8-block,True,31,fp8,https://huggingface.co/RedHatAI/gemma-4-31B-it-FP8-block,harbor,https://github.com/harbor-framework/harbor,500.0,12.0,0.606,525563275.0,0.0,4183675.0,529746950.0,166092.0,444527.0,52.73,1051126.0,0.0,8367.0,1059493.0,0.11,889.0,332.0
 SWE-Bench Verified,SWE-bench/SWE-bench_Verified,500,https://huggingface.co/datasets/SWE-bench/SWE-bench_Verified,Pi,[],True,https://github.com/earendil-works/pi/tree/main,Gemma4-31B-FP8,RedHatAI/gemma-4-31B-it-FP8-block,True,31,fp8,https://huggingface.co/RedHatAI/gemma-4-31B-it-FP8-block,harbor,https://github.com/harbor-framework/harbor,500.0,0.0,0.574,377479979.0,0.0,4177117.0,381657096.0,122274.0,385336.0,38.82,754959.0,0.0,8354.0,763314.0,0.08,770.0,244.0
+RH SWE-Bench,rounakbende10/rh-swe-bench,357,https://huggingface.co/datasets/rounakbende/rh-swe-bench,Claude Code,[],False,https://github.com/anthropics/claude-code,Nemotron-3-Super-120B-NVFP4,RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4,True,120,nvfp4,https://huggingface.co/RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4,harbor,https://github.com/harbor-framework/harbor,357.0,14.0,0.224,783670817.0,0.0,5510291.0,789181108.0,144352.0,355122.0,38.03,2195156.0,0.0,15434.0,2210591.0,0.11,994.0,404.0
 RH SWE-Bench,rounakbende10/rh-swe-bench,357,https://huggingface.co/datasets/rounakbende/rh-swe-bench,OpenCode,[],True,https://github.com/anomalyco/opencode,Nemotron-3-Super-120B-NVFP4,RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4,True,120,nvfp4,https://huggingface.co/RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4,harbor,https://github.com/harbor-framework/harbor,357.0,7.0,0.308,839586046.0,0.0,5706139.0,845292185.0,94762.0,251102.0,37.44,2351781.0,0.0,15983.0,2367765.0,0.1,703.0,265.0
 RH SWE-Bench,rounakbende10/rh-swe-bench,357,https://huggingface.co/datasets/rounakbende/rh-swe-bench,Pi,[],True,https://github.com/earendil-works/pi/tree/main,Nemotron-3-Super-120B-NVFP4,RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4,True,120,nvfp4,https://huggingface.co/RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4,harbor,https://github.com/harbor-framework/harbor,357.0,13.0,0.216,770574986.0,0.0,6519128.0,777094114.0,116348.0,360934.0,30.65,2158473.0,0.0,18260.0,2176734.0,0.09,1011.0,325.0
+SWE-Bench Pro -- Ansible,ScaleAI/SWE-bench_Pro,96,https://huggingface.co/datasets/ScaleAI/SWE-bench_Pro,Claude Code,[],False,https://github.com/anthropics/claude-code,Nemotron-3-Super-120B-NVFP4,RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4,True,120,nvfp4,https://huggingface.co/RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4,harbor,https://github.com/harbor-framework/harbor,96.0,0.0,0.432,406026150.0,0.0,2839236.0,408865386.0,100355.0,115220.0,26.44,4229439.0,0.0,29575.0,4259014.0,0.28,1200.0,1045.0
 SWE-Bench Pro -- Ansible,ScaleAI/SWE-bench_Pro,96,https://huggingface.co/datasets/ScaleAI/SWE-bench_Pro,OpenCode,[],True,https://github.com/anomalyco/opencode,Nemotron-3-Super-120B-NVFP4,RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4,True,120,nvfp4,https://huggingface.co/RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4,harbor,https://github.com/harbor-framework/harbor,96.0,4.0,0.323,686392098.0,0.0,2915876.0,689307974.0,86130.0,99705.0,19.45,7149917.0,0.0,30373.0,7180291.0,0.2,1038.0,897.0
 SWE-Bench Pro -- Ansible,ScaleAI/SWE-bench_Pro,96,https://huggingface.co/datasets/ScaleAI/SWE-bench_Pro,Pi,[],True,https://github.com/earendil-works/pi/tree/main,Nemotron-3-Super-120B-NVFP4,RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4,True,120,nvfp4,https://huggingface.co/RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4,harbor,https://github.com/harbor-framework/harbor,96.0,14.0,0.375,1051387065.0,0.0,3651367.0,1055038432.0,113768.0,127326.0,25.69,10951948.0,0.0,38035.0,10989983.0,0.27,1326.0,1185.0
+SWE-Bench Verified,SWE-bench/SWE-bench_Verified,500,https://huggingface.co/datasets/SWE-bench/SWE-bench_Verified,Claude Code,[],False,https://github.com/anthropics/claude-code,Nemotron-3-Super-120B-NVFP4,RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4,True,120,nvfp4,https://huggingface.co/RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4,harbor,https://github.com/harbor-framework/harbor,500.0,0.0,0.472,882330271.0,0.0,6689285.0,889019556.0,174288.0,410828.0,45.91,1764660.0,0.0,13378.0,1778039.0,0.09,821.0,348.0
 SWE-Bench Verified,SWE-bench/SWE-bench_Verified,500,https://huggingface.co/datasets/SWE-bench/SWE-bench_Verified,OpenCode,[],True,https://github.com/anomalyco/opencode,Nemotron-3-Super-120B-NVFP4,RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4,True,120,nvfp4,https://huggingface.co/RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4,harbor,https://github.com/harbor-framework/harbor,500.0,1.0,0.41,724984042.0,0.0,5941303.0,730925345.0,93088.0,518565.0,21.02,1449968.0,0.0,11882.0,1461850.0,0.04,1037.0,186.0
 SWE-Bench Verified,SWE-bench/SWE-bench_Verified,500,https://huggingface.co/datasets/SWE-bench/SWE-bench_Verified,Pi,[],True,https://github.com/earendil-works/pi/tree/main,Nemotron-3-Super-120B-NVFP4,RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4,True,120,nvfp4,https://huggingface.co/RedHatAI/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4,harbor,https://github.com/harbor-framework/harbor,500.0,6.0,0.498,996839122.0,0.0,7196671.0,1004035793.0,144084.0,446845.0,32.53,1993678.0,0.0,14393.0,2008071.0,0.07,893.0,288.0
 RH SWE-Bench,rounakbende10/rh-swe-bench,357,https://huggingface.co/datasets/rounakbende/rh-swe-bench,Claude Code,[],False,https://github.com/anthropics/claude-code,Opus 4.6,Opus 4.6,False,1000,bf16,https://www.anthropic.com/news/claude-opus-4-6,harbor,https://github.com/harbor-framework/harbor,357.0,0.0,0.633,438501505.0,419608736.0,2823496.0,860933737.0,126529.0,243310.0,434.54,1228295.0,1175374.0,7908.0,2411579.0,1.22,681.0,354.0