bigcode-models-leaderboard / code_eval_board.csv
loubnabnl's picture
loubnabnl HF staff
update codegeex scores
ea10e12
raw
history blame
1.68 kB
Models,Size (B),Win Rate,Average score,Throughput (tokens/s),Seq_length,#Languages,humaneval-python,java,javascript,cpp,php,julia,d,lua,r,racket,rust,swift,Throughput (tokens/s) bs=50,Peak Memory (MB)
WizardCoder-15B-V1.0,15.0,11.92,31.43,43.7,8192,86,50.53,35.77,41.91,38.95,39.34,33.98,12.14,27.85,22.53,13.39,33.74,27.06,1470.0,32414
StarCoder-15B,15.0,9.88,22.74,43.9,8192,86,33.57,30.22,30.79,31.55,26.08,23.02,13.57,23.89,15.5,0.07,21.84,22.74,1490.0,33461
StarCoderBase-15B,15.0,9.83,22.4,43.8,8192,86,30.35,28.53,31.7,30.56,26.75,21.09,10.01,26.61,10.18,11.77,24.46,16.74,1460.0,32366
CodeGeex2-6B,6.0,8.58,21.23,32.7,8192,100,33.49,23.46,29.9,28.45,25.27,20.93,8.44,15.94,14.58,11.75,20.45,22.06,1100,14110
StarCoderBase-7B,7.0,8.08,20.17,46.9,8192,86,28.37,24.44,27.35,23.3,22.12,21.77,8.1,23.35,14.51,11.08,22.6,15.1,1700.0,16512
CodeGen25-7B-multi,7.0,7.96,20.04,32.6,2048,86,28.7,26.01,26.27,25.75,21.98,19.11,8.84,23.44,11.59,10.37,21.84,16.62,680.0,15336
StarCoderBase-3B,3.0,5.42,15.29,50.0,8192,86,21.5,19.25,21.32,19.43,18.55,16.1,4.97,18.04,10.1,7.87,16.32,9.98,1770.0,8414
Replit-2.7B,2.7,4.25,11.62,42.2,2048,20,20.12,21.39,20.18,20.37,16.14,1.24,6.41,2.11,7.2,3.22,15.19,5.88,577.0,7176
CodeGen25-7B-mono,7.0,4.08,12.1,34.1,2048,86,33.08,19.75,23.22,18.62,16.75,4.65,4.32,6.75,4.41,4.07,7.83,1.71,687.0,15336
CodeGen-16B-Multi,16.0,3.54,9.89,17.2,2048,6,19.26,22.2,19.15,21.0,8.37,0.0,7.68,8.5,6.45,0.66,4.21,1.25,0.0,32890
StarCoderBase-1.1B,1.1,3.17,9.81,71.4,8192,86,15.17,14.2,13.38,11.68,9.94,11.31,4.65,12.52,5.73,5.03,10.24,3.92,2360.0,4586
SantaCoder-1.1B,1.1,1.29,4.92,50.8,2048,3,18.12,15.0,15.47,6.2,1.5,0.0,0.0,0.1,0.0,0.0,2.0,0.7,2270.0,4602