AI 模型定价目录

GPT-5.3 Codex (xhigh) is amongst the leading models in intelligence, but somewhat expensive when comparing to other models of similar price. It's also faster than average, however very verbose. The model supports text and image input, outputs text, and has a 400k tokens context window with knowledge up to August 2025. GPT-5.3 Codex (xhigh) scores 54 on the Artificial Analysis Intelligence Index, placing it well above average among comparable models (averaging 36 ). When evaluating the Intelligence Index, it generated 77M tokens, which is very verbose in comparison to the average of 36M .

GPT-5.4 (xhigh) is amongst the leading models in intelligence, but somewhat expensive when comparing to other models of similar price. It's also faster than average, however very verbose. The model supports text and image input, outputs text, and has a 1m tokens context window with knowledge up to August 2025. GPT-5.4 (xhigh) scores 57 on the Artificial Analysis Intelligence Index, placing it well above average among comparable models (averaging 36 ). When evaluating the Intelligence Index, it generated 120M tokens, which is very verbose in comparison to the average of 36M .

GPT-5.4 mini (xhigh) is amongst the leading models in intelligence and well priced when comparing to other models of similar price. It's also notably fast, however very verbose. The model supports text and image input, outputs text, and has a 400k tokens context window with knowledge up to August 2025. GPT-5.4 mini (xhigh) scores 49 on the Artificial Analysis Intelligence Index, placing it well above average among comparable models (averaging 36 ). When evaluating the Intelligence Index, it generated 240M tokens, which is very verbose in comparison to the average of 36M .

GPT-5.5 (xhigh) is amongst the leading models in intelligence, but particularly expensive when comparing to other models of similar price. It's also faster than average, however somewhat verbose. The model supports text and image input, outputs text, and has a 922k tokens context window. GPT-5.5 (xhigh) scores 60 on the Artificial Analysis Intelligence Index, placing it well above average among comparable models (averaging 36 ). When evaluating the Intelligence Index, it generated 75M tokens, which is somewhat verbose in comparison to the average of 36M .

Claude 4.5 Haiku (Non-reasoning) is above average in intelligence and well priced when comparing to other non-reasoning models of similar price. It's also notably fast, however somewhat verbose. The model supports text and image input, outputs text, and has a 200k tokens context window with knowledge up to July 2025. Claude 4.5 Haiku (Non-reasoning) scores 31 on the Artificial Analysis Intelligence Index, placing it above average among comparable models (averaging 24 ). When evaluating the Intelligence Index, it generated 8.3M tokens, which is somewhat verbose in comparison to the average of 7.9M .

Claude Opus 4.6 (Adaptive Reasoning, Max Effort) is amongst the leading models in intelligence, but particularly expensive when comparing to other models of similar price. It's also notably slow and very verbose. The model supports text and image input, outputs text, and has a 1m tokens context window. Claude Opus 4.6 (Adaptive Reasoning, Max Effort) scores 53 on the Artificial Analysis Intelligence Index, placing it well above average among comparable models (averaging 36 ). When evaluating the Intelligence Index, it generated 160M tokens, which is very verbose in comparison to the average of 36M .

Claude Opus 4.7 (Adaptive Reasoning, Max Effort) is amongst the leading models in intelligence, but particularly expensive when comparing to other models of similar price. It's also slower than average and very verbose. The model supports text and image input, outputs text, and has a 1m tokens context window with knowledge up to January 2026. Claude Opus 4.7 (Adaptive Reasoning, Max Effort) scores 57 on the Artificial Analysis Intelligence Index, placing it well above average among comparable models (averaging 36 ). When evaluating the Intelligence Index, it generated 110M tokens, which is very verbose in comparison to the average of 36M .

Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) is amongst the leading models in intelligence, but somewhat expensive when comparing to other models of similar price. It's also slower than average and very verbose. The model supports text and image input, outputs text, and has a 1m tokens context window. Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) scores 52 on the Artificial Analysis Intelligence Index, placing it well above average among comparable models (averaging 36 ). When evaluating the Intelligence Index, it generated 200M tokens, which is very verbose in comparison to the average of 36M .

Gemini 3.1 Pro Preview is amongst the leading models in intelligence, but somewhat expensive when comparing to other models of similar price. It's also notably fast, however somewhat verbose. The model supports text, image, speech, and video input, outputs text, and has a 1m tokens context window. Gemini 3.1 Pro Preview scores 57 on the Artificial Analysis Intelligence Index, placing it well above average among comparable models (averaging 36 ). When evaluating the Intelligence Index, it generated 57M tokens, which is somewhat verbose in comparison to the average of 36M .

Gemini 3.5 Flash (high) is amongst the leading models in intelligence and reasonably priced when comparing to other models of similar price. It's also notably fast, however somewhat verbose. The model supports text, image, speech, and video input, outputs text, and has a 1m tokens context window. Gemini 3.5 Flash (high) scores 55 on the Artificial Analysis Intelligence Index, placing it well above average among comparable models (averaging 36 ). When evaluating the Intelligence Index, it generated 73M tokens, which is somewhat verbose in comparison to the average of 36M .