Unverified50% confidenceFactExact time
Qwen3 在训练时支持长达 128K tokens 的上下文
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
8/18/2026
First Seen
Valid until: 11/16/2026
Sources
Related Claims
UnverifiedQwen3.7 Max's context window reaches 256K tokens.70% similarUnverifiedEarly LLaMA-2 only supported a 4K token context, whereas new-generation models such as DeepSeek-V3 and Qwen2.5 have extended the context to 128K or even longer.69% similarUnverifiedQwen-2引入了GQA、支持128K token的上下文窗口和基于字节级BPE的分词器（词表约15万），参数规模从0.5B到72B69% similarUnverifiedQwen3-27B natively supports a long context of approximately 260,000 tokens, which can be extended to the 1,000,000-token scale through technologies such as YaRN.68% similarVerifiedThe Qwen series models support context windows ranging from 8K to 128K.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/771246API
curl https://kongchang.com/api/v1/knowledge/claims/771246MCP
get_claim(id=771246)