Unverified60% confidenceFactTime unknown
RTX 4090的理论FP16算力为82.6 TFLOPS，显存带宽为1008 GB/s
2
Sources
60%
Confidence
Long-term
Relevance
6/1/2026
First Seen
Sources
Qwen3.6 MTP加速实测：单GPU推理飙到220 token/s
bilibili普通鱼学家呀
Related Entities
Related Claims
UnverifiedRTX 4090的GDDR6X显存接口带宽高达1008 GB/s83% similarPartially VerifiedA100 80GB采用HBM2e显存，带宽约2039GB/s；H100 SXM采用HBM3，带宽约3.35TB/s，FP64 Tensor Core算力约67 TFLOPS，是A100（约19.5 TFLOPS）的3倍以上80% similarVerifiedRTX A6000基于Ampere架构，搭载10496个CUDA核心与48GB GDDR6显存，带宽768 GB/s，FP32算力38.7 TFLOPS78% similarVerified主流RTX 4090显卡仅有24GB显存，A100/H100专业级显卡每张80GB78% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/18050API
curl https://kongchang.com/api/v1/knowledge/claims/18050MCP
get_claim(id=18050)