Unverified50% confidenceBenchmarkExact time
An 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.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
8/19/2026
First Seen
Valid until: 11/17/2026
Sources
Related Claims
UnverifiedA model with 70 billion parameters (70B) can run on a single RTX 4090 (24GB VRAM) after quantization.71% similarUnverifiedIt has become possible to fine-tune 70B-class models on a single A100 80GB GPU using QLoRA.66% similarVerifiedQLoRA introduces 4-bit quantization technology, making it possible to fine-tune models with 65 billion parameters on a single RTX 3090/4090 GPU with 24 GB of VRAM.65% similarUnverifiedMiniCPM和Qwen-0.5B等1B以下轻量模型配合INT4量化可以在手机芯片上运行65% similarUnverified原本需要 A100 80GB 才能完成的 70B 参数模型微调任务，在 Unsloth 优化下可以在消费级 RTX 4090（24GB）上运行65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/775583API
curl https://kongchang.com/api/v1/knowledge/claims/775583MCP
get_claim(id=775583)