Unverified50% confidenceOpinionExact time
Download counts for Flash lightweight models are generally higher than those for flagship versions, reflecting users' greater preference for models that can run on local GPUs.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/9/2026
First Seen
Valid until: 12/8/2026
Sources
Related Entities
Related Claims
UnverifiedFlash-series models typically adopt a parameter scale of under 70B.68% similarUnverifiedThe DeepSeek Flash model can output more than 100 tokens per second.67% similarUnverifiedBecause Flash models have fast inference speeds (50–200ms per request vs. 500–2000ms for flagship models), they can support higher QPS under the same TPM limits.64% similarUnverifiedUnder certain configurations, graphics cards with 16GB of VRAM can also complete video generation with the Wan model at lower resolutions.64% similarUnverified盘古2.0 Flash在LiveCodeBench算法推理测试中得分85.1，仅次于DeepSeek V4 Flash的91.6，超过千问3.6 27B的83.963% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/885796API
curl https://kongchang.com/api/v1/knowledge/claims/885796MCP
get_claim(id=885796)