Unverified50% confidenceFactExact time
The public API of DeepEP Ascend is fully aligned with the NVIDIA version of DeepEP, including EP Buffer, Dispatch, Combine, and the defer epilogue parameter, meaning code running on NVIDIA cards requires virtually no changes at the Python layer after being ported to Ascend.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/3/2026
First Seen
Valid until: 1/1/2027
Sources
Related Entities
Related Claims
UnverifiedWhat is truly ready in DeepEP Ascend at present is the EP cross-card dispatch mechanism; APIs for other workload strategies such as PP, bucket, and ngram have been placeholder-defined but their implementations are not yet complete.71% similarVerifiedThe NVIDIA DeepStream SDK is built on the GStreamer multimedia framework, organizes processing nodes in the form of plugins, and passes data through GPU memory in a zero-copy manner.64% similarUnverifiedDeepSeek ported the DeepEP communication library to the domestically produced Ascend NPU and open-sourced it on GitHub under the name DeepEP Ascend.62% similarUnverifiedThe Ascend version of DeepEP must use the expanded layout (Dispatch Expand) and currently operates as a single communication domain.61% similarUnverifiedThe DeepSeek community has open-sourced two underlying acceleration projects targeting the Ascend platform: DeepEP Ascend and DeepGEMM Ascend60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/970615API
curl https://kongchang.com/api/v1/knowledge/claims/970615MCP
get_claim(id=970615)