lu

w2 start -S /home/www/whistle/8888 -C -p 8888 w2 start -S /home/www/whistle/9999 -C -p 9999 通过上面命令起两个实例其中一个会过一段时间挂掉

> Renamed the issue. Let me know if this is accurate: we need an option to entirely disable the request to huggingface and only load from local disk. If it...

> What device? mali gpu arm64

> Are you using a terminal emulator? Which device? Which version of android? termux xiaomi 14pro android 14

proot-distro installing debian can run exo, but when multiple phones are connected to the same network, exo cannot be added together, and all nodes are displayed as one.

可以硬编码模型路径修改三个地方： exo/exo/api/chatgpt_api.py: resolve_tinygrad_tokenizer function def resolve_tinygrad_tokenizer(model_id: str): if model_id == "llama3-8b-sfr": # here i modified # return AutoTokenizer.from_pretrained("TriAiExperiments/SFR-Iterative-DPO-LLaMA-3-8B-R") return AutoTokenizer.from_pretrained("/nasroot/models/Meta-Llama-3-8B") elif model_id == "llama3-70b-sfr": return AutoTokenizer.from_pretrained("TriAiExperiments/SFR-Iterative-DPO-LLaMA-3-8B-R") else: raise ValueError(f"tinygrad doesnt...

> @artistlu 感谢您的建议，我依旧没有成功，我无法连接到 huggingface，未能成功运行该项目 我也不能链接 huggingface，我是下在这[https://www.modelscope.cn/models/LLM-Research/Meta-Llama-3-8B/files](url) 下载模型，放在每个节点固定目录，硬编码三个地方就可以加载了。   希望有帮助

> Try running with `SUPPORT_BF16=0` e.g. `SUPPORT_BF16=0 python3 main.py`. Can you let me know if that works? > > Ideally we detect this automatically. In order to load a local...