Taobao MNN team open sources 0.3B mobile model
The story so far
Taobao's MNN team has open-sourced a 0.3B mobile model MNN-Pocket-0.3B, which can run at 350 tokens/s on a real iPhone 18 Pro offline. The model's peak memory is below 300MB, decoding throughput exceeds 200 tokens/s, INT4 deployment file is 173MB, and the English IFEval score is 73.20, while the Chinese instruction following IFEval-zh score is 74.59. This model is suitable for mobile deployment, and Taobao's MNN team claims that its goal is to lower the running threshold of large models on mobile phones.
AI-written from the reports · Updated 13 days ago
Coverage timeline
Follow the coverage to see every side of the story.
- ITHome · AITaobao MNN team open-sources 0.3B mobile model MNN-Pocket-0.3B
Taobao MNN team open-sources 0.3B mobile model MNN-Pocket-0.3B, achieving 350 tokens/s on real iPhone 18 Pro offline. The model's peak memory is below 300MB, decoding throughput exceeds 200 tokens/s, INT4 deployment file is 173MB, English IFEval score is 73.20, and Chinese instruction following IFEval-zh score is 74.59.