Unverified50% confidenceFactExact time
早期模型的上下文窗口限制为4K tokens
1
Sources
50%
Confidence
Long-term
Relevance
7/22/2026
First Seen
Sources
Related Claims
UnverifiedEarly models had relatively small context windows, approximately 4K–8K tokens.82% similarUnverifiedEarly Codex had a context window of only approximately 4K tokens, allowing it to handle only single-file-level code comprehension tasks.81% similarUnverifiedLLM上下文窗口从早期的4K token扩展到Claude模型的200K token78% similarUnverifiedMany free or low-cost large language models may have context windows of only 4K–8K tokens.76% similarVerifiedEarly large language models only supported context windows of 4K–8K tokens, while current mainstream models have expanded to 100K or even 200K tokens.75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/587582API
curl https://kongchang.com/api/v1/knowledge/claims/587582MCP
get_claim(id=587582)