Unverified50% confidenceBenchmarkExact time
The Huawei Ascend 950PR successfully ran a pruned version of the DeepSeek V4-Flash model in a single-card configuration, achieving a peak decode speed of 208 TPS.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/28/2026
First Seen
Valid until: 12/27/2026
Sources
Related Entities
Related Claims
UnverifiedA single TPU v4 chip has a peak compute of approximately 275 TFLOPS (BF16 precision), and a TPU v4-32 pod consists of 32 chips interconnected via ICI high-speed interconnects, with a total compute of approximately 8,800 TFLOPS.71% similarUnverifiedThe Huawei Ascend 910B has single-card compute performance approaching the level of the A100, but there is a significant gap with NVIDIA in multi-card interconnect efficiency.71% similarUnverifiedThe UD-Q8_K_XL quantized version of DeepSeek V4 Flash has a file size of 162 GB, uses MXFP4+BF16 mixed precision, and achieves completely lossless quality.71% similarUnverifiedHuawei's Ascend 950 super-node physical machine has officially made its debut, achieving 1,024-card interconnection with computing power reaching the exaflops scale.70% similarUnverifiedOn a 128G compute module device, the decoding speed of Qwen3.8 Flash routinely reaches 241 TPS, with a peak of 274 TPS.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/956037API
curl https://kongchang.com/api/v1/knowledge/claims/956037MCP
get_claim(id=956037)