Unverified50% confidenceFactExact time
DeepInfer's self-developed optimized kernels include Flash MLA, Deep GMM, and Deep EP.
1
Sources
50%
Confidence
Long-term
Relevance
10/25/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedDeepSeek is deploying native multimodal capabilities at the Flash level rather than the flagship version first, with the intent to prioritize capturing high-frequency, large-scale application scenarios.80% similarVerifiedDeepSeek adopts a Mixture of Experts (MoE) architecture and a Multi-head Latent Attention (MLA) mechanism.76% similarUnverifiedDeepSeek kernel engineers wrote the core attention kernel for DeepSeek V4.1 Flash.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1095878API
curl https://kongchang.com/api/v1/knowledge/claims/1095878MCP
get_claim(id=1095878)