Unverified50% confidenceBenchmarkExact time
The MegaMoE operator achieves a hardware-level performance of 852 TFLOPS
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/1/2026
First Seen
Valid until: 12/30/2026
Sources
Related Entities
Related Claims
UnverifiedDeepSeek published a sparse prefill figure of 1350 TFLOPS (absolute value) for the NVIDIA B200, which uses a different metric from the 95% peak utilization figure cited for Ascend.66% similarUnverifiedFlashMLA's prefill throughput for the sparse attention operator on the Ascend 950 reaches 410 TFLOPS (95% of the hardware's theoretical peak), and decode reaches 360 TFLOPS (83%), both figures as published by DeepSeek itself with no third-party measurements.64% similarUnverifiedAt a 4K×4K×16K matrix multiplication scale, the M5 Ultra achieved 91 TFLOPS, while the M5 Max and M3 Ultra achieved 34 and 22 TFLOPS respectively.63% similarUnverifiedA 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.63% similarVerifiedRTX 4090的理论FP16算力为82.6 TFLOPS，显存带宽为1008 GB/s63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/963585API
curl https://kongchang.com/api/v1/knowledge/claims/963585MCP
get_claim(id=963585)