Unverified50% confidenceFactExact time
英伟达A100单卡算力约312 TFLOPS（FP16）
1
Sources
50%
Confidence
Long-term
Relevance
7/9/2026
First Seen
Sources
GPT-5.6三模型详解：Sol、Terra、Luna分层定位与能力对比
bilibiliX信息差6/28/2026
Related Claims
UnverifiedT4 GPU拥有16GB显存和8.1 TFLOPS的FP32算力74% similarVerifiedNVIDIA A100单卡拥有80GB HBM2e显存和312 TFLOPS的FP16算力73% similarUnverifiedTPU v4单芯片峰值算力约275 TFLOPS（BF16精度），TPU v4-32 pod由32个芯片通过ICI高速互联组成，总算力约8800 TFLOPS73% similarUnverifiedNVIDIA H100的屋脊点约为597 FLOP/Byte（约2000 TFLOPS ÷ 3.35 TB/s）72% similarVerifiedNVIDIA H100单卡FP64双精度浮点性能达到约34 TFLOPS71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/401452API
curl https://kongchang.com/api/v1/knowledge/claims/401452MCP
get_claim(id=401452)