Unverified50% confidenceFactExact time
The generation speed of cloud-based APIs such as ChatGPT is typically 30–80 tok/s, while locally running small 7B models on Apple Silicon can achieve 40–60 tok/s.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/3/2026
First Seen
Valid until: 12/2/2026
Sources
Related Entities
Related Claims
Unverified现代云内网 RTT 可低至 0.1–0.5 毫秒，与本地 SSD 的 fsync 延迟处于同一量级66% similarUnverified运行Stable Diffusion + ControlNet生成艺术二维码建议使用8GB以上显存的GPU（如NVIDIA RTX 3060及以上），Apple Silicon可通过MPS后端运行但速度较慢65% similarUnverifiedGPT-OSS 20B在RTX 3090上可达约40 Token/s的生成速度，在RTX 5090上能达到200 Token/s65% similarUnverifiedOn the Intel Skylake microarchitecture, the latency of 64-bit IDIV is 35–88 cycles.64% similarUnverifiedA ~25 GB Q3 model running entirely within the RTX 4090's VRAM (bandwidth ~1 TB/s) theoretically achieves ~40 tokens per second, whereas running it in DDR5 memory (~80 GB/s) drops throughput to 1–3 tokens per second.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/848676API
curl https://kongchang.com/api/v1/knowledge/claims/848676MCP
get_claim(id=848676)