Unverified50% confidenceFactExact time
Since 2023, the NPU computing power of flagship smartphone chips has exceeded 40 TOPS, and combined with quantization techniques, 7B and even 13B parameter models can be inferred on smartphones at a speed of several tokens per second.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
8/4/2026
First Seen
Valid until: 11/2/2026
Sources
Nightcrawler：手机本地运行的AI渗透测试代理深度解析
Related Claims
UnverifiedHuawei Cloud's ASS zero-zone computing cluster supports 100,000-card-scale computing power, compresses Token generation latency to within 10 milliseconds, and achieves computing power of 2001 PFLOPS.66% similarUnverifiedOn high-performance laptops, quantized Gemma models can achieve token generation speeds of 20–50 tokens per second; pure CPU inference is approximately 5–15 tokens/second, and INT4 quantization on NVIDIA RTX 4060/4070 can boost inference speed by 3–5 times.65% similarUnverifiedAn INT8-quantized MobileNet-v2 can complete a single image classification inference on an STM32H7 in approximately 100ms, with power consumption on the order of hundreds of milliwatts.64% similarUnverifiedRunning the Qwen3.6 27B model on a GPU with 24 GB of VRAM, the generation speed reaches approximately 40 tokens per second without optimization, and 50–60 tokens/s after optimization.63% similarUnverifiedNVIDIA芯片从A100（7nm，400W TDP）到H100（4nm，700W TDP，推理吞吐提升约6倍）再到B200（4nm，1000W TDP，性能再提升约4倍）63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/684157API
curl https://kongchang.com/api/v1/knowledge/claims/684157MCP
get_claim(id=684157)