MiniCPM
MiniCPM copied to clipboard
OpenBMB
MiniCPM4: Ultra-Efficient LLMs on End Devices, achieving 5+ speedup on typical end-side chips
### Feature request / 功能建议 仓库地址: https://github.com/ztxz16/fastllm 类似 llama.cpp，对国产化模型的支持程度更好，且运行速度更快。 >fastllm分别实现了llm中常见的算子，并尽可能将一个算子合为一个CUDA kernel，只有一轮GPU核心到显存的输入输出，提高了计算访存比。
### Description / 描述 當於LM Studio載入"minicpm-2b-dpo-fp32.Q6_K.gguf" 時，報錯："create_tensor: tensor 'output.weight' not found"，不知道應該設定甚麼Preset ### Case Explaination / 案例解释 _No response_
Hi I triied to convert vllm ``` python inference/convert_hf_to_vllmcpm.py --load ./cache/MiniCPM-V/ --save ./cache/MiniCPM-V-vllm ``` But got the following error: FileNotFoundError: [Errno 2] No such file or directory: '.cache/MiniCPM-V/pytorch_model.bin' Thanks
### Is there an existing issue ? / 是否已有相关的 issue ? - [X] I have searched, and there is no existing issue. / 我已经搜索过了，没有相关的 issue。 ### Describe the bug /...
### Feature request / 功能建议 Hello authors, thank you for sharing such a powerful LLM with the community! I was wondering if it's possible to make the script of converting...
### Feature request / 功能建议 使用LoRa技术微调，产生了下面的这些文件，有没有使用微调后的模型和原始模型一起进行预测的代码呢？ 
### Is there an existing issue ? / 是否已有相关的 issue ? - [X] I have searched, and there is no existing issue. / 我已经搜索过了，没有相关的 issue。 ### Describe the bug /...
### Feature request / 功能建议 T.D.
### Description / 描述 目前我使用Langchain试图进行多轮对话时提示Requested tokens (939) exceed context window of 512，请问有没有办法调整这个窗口值呢 ### Case Explaination / 案例解释 _No response_
In your blog you bring many runs and iterations and scaling laws. Could you share the training losses and parameters from which those graphs are made for further research and...
