Expired85% confidenceFactTime unknown
QwQ-32B performs nearly on par with the full version of DeepSeek R1 across multiple mainstream benchmarks, and surpasses it on some tasks.
1
Sources
85%
Confidence
Medium-term (~90 days)
Relevance
6/1/2026
First Seen
Valid until: 8/30/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedBenchmark tests show that Qwen3-235B's performance matches or even surpasses DeepSeek R1, with advantages in categories such as math, coding, and reasoning.77% similarUnverifiedOn code-related tasks such as LiveBench, QwQ-32B surpasses DeepSeek R1 by approximately two percentage points.76% similarUnverifiedQwen2.5-72B approaches the performance of Llama-3.1-405B across multiple benchmarks.68% similarUnverifiedQwen3.6-27B's benchmark scores place it at the leading position among models of similar scale (20B–30B parameter range).66% similarUnverifiedReddit社区的大规模基准测试基于BeeLlama.cpp v0.4.0，在Qwen 3.6 27B和Gemma 4 31B两款模型上累计跑了413组配置对比，其中Qwen模型238组、Gemma模型175组66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/23192API
curl https://kongchang.com/api/v1/knowledge/claims/23192MCP
get_claim(id=23192)