Unverified50% confidenceFactExact time
The qwen3.5:122b tag requires 81 GB of VRAM and is used for deeper reasoning.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/12/2026
First Seen
Valid until: 1/10/2027
Sources
Related Entities
Related Claims
UnverifiedQwen-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% similarUnverifiedQwen3.8-Flash-Next has a total parameter count of 125B with approximately 6B activated parameters, and belongs to the sparse MoE architecture.69% similarUnverifiedQwen3.5-122B-A10B is a MoE architecture model with approximately 122B total parameters and approximately 10B activated parameters.68% similarUnverifiedQwen3.8-Max-0902 uses a Mixture-of-Experts (MoE) architecture with 2.4T parameters and supports a context length of 128K.68% similarUnverifiedQwen3 27B has entered the Ollama official library and can be retrieved with ollama pull qwen3:27b; the default tag is approximately 18 GB, supports a 256k context, and retains text and image capabilities.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1001692API
curl https://kongchang.com/api/v1/knowledge/claims/1001692MCP
get_claim(id=1001692)