Unverified50% confidenceFactExact time
DeepInfer is released under the Apache 2.0 license, supports NVIDIA Blackwell and Huawei Ascend 950, and includes components such as expert load balancing, speculative decoding, and asynchronous scheduling.
1
Sources
50%
Confidence
Long-term
Relevance
10/21/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedDeepSeek open-sourced the inference framework DeepInfer on October 20, providing optimized operators, multi-card parallelism and asynchronous scheduling capabilities for V4.1 Flash, under the Apache 2.0 license.79% similarUnverifiedWhat DeepInfer open-sourced is the offline inference core; the complete online service chain still requires filling in the caching layer and load balancing yourself.72% similarUnverifiedDeepSeek open-sourced operators and communication components for Huawei Ascend, corresponding one-to-one with the previously released GPU versions.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1081843API
curl https://kongchang.com/api/v1/knowledge/claims/1081843MCP
get_claim(id=1081843)