Unverified50% confidenceFactExact time
以 GPT-4o 为例，一张 1024×1024 图片约消耗 765 个 token
1
Sources
50%
Confidence
Long-term
Relevance
8/27/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedGPT-5.4 introduces for the first time two image input detail levels, Original and High, with Original mode supporting a maximum of 10.24 million total pixels or a maximum single-side length of 6,000 pixels.75% similarVerifiedIn a test building an image studio application, the standard GPT-5 used approximately 23.6K tokens while GPT-5 Codex used approximately 27.8K tokens.73% similarVerifiedGPT-4 uses a vocabulary of approximately 100,000 tokens.72% similarUnverifiedGPT-4's single output limit is approximately 4,096–8,192 tokens (roughly 2,000–4,000 Chinese characters), and the input context window has been expanded to 128K or even longer71% similarUnverifiedGPT-4 has a context window of 128K, and a single request can theoretically consume tokens worth nearly $4.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/807694API
curl https://kongchang.com/api/v1/knowledge/claims/807694MCP
get_claim(id=807694)