Unverified60% confidenceBenchmarkExact time
On the 70-billion-parameter DeepSeek R1 distillation model, because the model overflows the RTX 5090's 32 GB of VRAM, the M5 Ultra is more than 20 times faster than the 5090.
2
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
9/23/2026
First Seen
Valid until: 12/22/2026
Sources
Related Entities
Related Claims
UnverifiedUnder a 70-billion-parameter workload, the 5090 draws approximately 275W and the M5 Ultra approximately 300W, making power consumption similar, yet the M5 Ultra is more than 20 times faster.76% similarUnverifiedLiquid AI d1-omni-600M 参数规模为 587M，Q8_0 GGUF 量化后约 670 MB，llama.cpp 已合并支持，可在树莓派 5 上通过 llama-server 命令运行。70% similarUnverifiedThe reference model Supra-50M has 51.8M parameters, is 103.6 MB in bf16, 56.2 MB after 8-bit quantization, perplexity degrades from 165 to 193 when reduced to int4, and collapses entirely when compressed to ternary.69% similarUnverifiedMixtral 8x7B has approximately 46.7 billion total parameters, but only activates approximately 13 billion parameters per forward pass; its inference compute is close to that of a 13B model, yet it still requires approximately 90 GB of VRAM (FP16).68% similarUnverified在M2 Ultra（最高192GB统一内存）上，MLX可流畅运行700亿参数量级的模型68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/944889API
curl https://kongchang.com/api/v1/knowledge/claims/944889MCP
get_claim(id=944889)