LLM API比較

ModelContext/OutputKnowledge Cutoff$Input/1M$Output/1M
o1-2024-12-17200k/100k2023-101560
o1-preview-2024-09-12128k/32k2023-101560
o1-mini-2024-09-12128k/64k2023-10312
gpt-4o-2024-11-20128k/16k2023-102.510
gpt-4o-mini-2024-07-18128k/16k2023-100.150.60
claude-3-5-sonnet-20241022200k/8k2024-04315
claude-3-5-haiku-20241022200k/8k2024-0715
claude-3-opus-20240229200k/4k2023-081575
claude-3-sonnet-20240229200k/4k2023-08315
claude-3-haiku-20240307200k/4k2023-080.251.25
gemini-2.0-flash-exp1M/8k2024-08?--
gemini-1.5-pro1M(2M)/8kearly 20230 (1.25/2.5*)0 (5-10*)
gemini-1.5-flash1M/8kearly 20230 (0.075-0.15*)0 (0.30-0.60*)
grok-beta128k515
* >128k tokens