Unverified50% confidenceFactExact time
The Neural Accelerator requires supporting software to function; running M5 Max with a 2023 old version of llama.cpp yields a read speed of only approximately 990 tokens/s.
1
Sources
50%
Confidence
Long-term
Relevance
10/5/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn actual benchmark threads for llama.cpp, the 2021 M1 Max achieved a decode write speed of 61 tokens/s, higher than the 2024 M4 Pro's 51 tokens/s.76% similarUnverified缺少独立 GPU 的设备运行 2B 参数模型时生成速度可能低于每秒 5 个 token67% similarUnverified一个800万参数（8.3M）的沙盒引擎在2019年发布的6核桌面CPU上跑出约2000 tokens/秒的解码速度67% similarUnverifieddecode阶段的生成速度近似等于内存带宽÷模型大小，例如在内存带宽273GB/s的M4 Pro上运行4GB的Q4量化7B模型理论最大生成速度约68 tokens/s67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/977843API
curl https://kongchang.com/api/v1/knowledge/claims/977843MCP
get_claim(id=977843)