Unverified50% confidenceFactExact time
The Qwen family offers the widest range of sizes, from sub-billion to tens of billions of parameters and beyond, supports text and image input, and provides compact quantized downloads.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/28/2026
First Seen
Valid until: 12/27/2026
Sources
Related Entities
Related Claims
UnverifiedQwen3 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.71% similarUnverifiedLlama, Qwen, and DeepSeek all offer versions at various parameter scales ranging from 1.5B to over 70B68% similarUnverifiedQwen2.5-Max-0902 supports a context window of 1 million tokens.68% similarUnverifiedFor users with 12 GB of VRAM, Q3_XXS remains viable, and its coding capability is still superior to Qwen3.6.67% similarUnverifiedQwen 3.6 35B A3B is fully open-source, with weights available on Hugging Face at approximately 72 GB in size.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/956124API
curl https://kongchang.com/api/v1/knowledge/claims/956124MCP
get_claim(id=956124)