empirically_measured_model_speeds.json•2.89 kB
{
"claude-3-5-haiku-20241022": {
"total_time_s": 9.66,
"output_tokens": 452,
"output_tokens_per_second": 46.79
},
"claude-3-7-sonnet-20250219": {
"total_time_s": 8.907,
"output_tokens": 469,
"output_tokens_per_second": 52.66
},
"claude-3-opus-20240229": {
"total_time_s": 20.186,
"output_tokens": 583,
"output_tokens_per_second": 28.88
},
"claude-3-sonnet-20240229": {
"total_time_s": 9.072,
"output_tokens": 518,
"output_tokens_per_second": 57.1
},
"deepseek-chat": {
"total_time_s": 21.641,
"output_tokens": 442,
"output_tokens_per_second": 20.42
},
"deepseek-reasoner": {
"total_time_s": 39.793,
"output_tokens": 1097,
"output_tokens_per_second": 27.57
},
"gemini-2.0-flash": {
"total_time_s": 2.59,
"output_tokens": 488,
"output_tokens_per_second": 188.44
},
"gemini-2.0-flash-lite": {
"total_time_s": 2.93,
"output_tokens": 467,
"output_tokens_per_second": 159.4
},
"gemini-2.0-flash-thinking-exp-01-21": {
"total_time_s": 6.103,
"output_tokens": 460,
"output_tokens_per_second": 75.37
},
"gemini-2.5-pro-preview-03-25": {
"total_time_s": 18.481,
"output_tokens": 593,
"output_tokens_per_second": 32.09
},
"gpt-4.1": {
"total_time_s": 8.908,
"output_tokens": 441,
"output_tokens_per_second": 49.51
},
"gpt-4.1-mini": {
"total_time_s": 6.184,
"output_tokens": 506,
"output_tokens_per_second": 81.82
},
"gpt-4.1-nano": {
"total_time_s": 3.315,
"output_tokens": 434,
"output_tokens_per_second": 130.92
},
"gpt-4o": {
"total_time_s": 10.358,
"output_tokens": 446,
"output_tokens_per_second": 43.06
},
"gpt-4o-mini": {
"total_time_s": 6.086,
"output_tokens": 416,
"output_tokens_per_second": 68.36
},
"grok-3-fast-latest": {
"total_time_s": 5.915,
"output_tokens": 396,
"output_tokens_per_second": 66.95
},
"grok-3-latest": {
"total_time_s": 13.587,
"output_tokens": 396,
"output_tokens_per_second": 29.15
},
"grok-3-mini-fast-latest": {
"total_time_s": 6.382,
"output_tokens": 345,
"output_tokens_per_second": 54.06
},
"grok-3-mini-latest": {
"total_time_s": 9.32,
"output_tokens": 366,
"output_tokens_per_second": 39.27
},
"o1-preview": {
"total_time_s": 19.538,
"output_tokens": 1394,
"output_tokens_per_second": 71.35
},
"o3-mini": {
"total_time_s": 5.416,
"output_tokens": 395,
"output_tokens_per_second": 72.93
}
}