GSM-Infinite-Leaderboard / data /long_context.csv
atlas5301
initial commit
87248a0
raw
history blame
491 Bytes
Model,8K,16K,32K,Average↑
gemini-1.5-pro-002,1182.43,896.31,812.96,963.9
qwen-2.5-72b-instruct,927.33,681.53,563.65,724.17
mistral-large-2411,914.49,563.73,319.21,599.14
deepseek-v3,935.10,477.02,313.66,575.2
gemini-1.5-flash-002,673.88,476.72,377.38,509.3
llama-3.1-70b-instruct,479.00,394.50,355.5,409.67
minimax-text-01,481.32,359.56,325.95,388.94
gpt-4o-mini,401.00,337.81,275.63,338.15
qwen-2.5-7b-instruct,248.00,211.50,196.17,218.56
llama-3.1-8b-instruct,183.67,149.50,109.45,147.54