Unverified50% confidenceReleaseExact time
This FlashMLA release is a breaking change: it removes support for the Hopper architecture and V3, V3.2, V4.0, and changes the KV Cache format for FP8 and FP4, making it incompatible with previous versions.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/1/2026
First Seen
Valid until: 12/30/2026
Sources
Related Entities
Related Claims
UnverifiedPangu 2.0 Flash currently does not support multimodality.72% similarUnverifiedQwen3.8 Flash is divided into two products: the open-weight version Qwen3.8 Flash Next and the cloud production version Qwen3.8 Flash.66% similarUnverifiedSGLang v0.5.18 upgrades its dependencies to PyTorch 2.13.0 (with triton 3.7.1), FlashInfer 0.6.17, CuTeDSL 4.6.2, and sgl-kernel 0.4.6.post1.64% similarUnverifiedThe official new 3.8 Flash model has been released.62% similarUnverifiedIn the 'Xianxia Milk Tea Shop' long-horizon task test, MiMo-V2.6 Pro encountered a 404 error, delivered a thinner implementation than its own Flash version, and took twice as long as Flash.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/963643API
curl https://kongchang.com/api/v1/knowledge/claims/963643MCP
get_claim(id=963643)