JocelynPanPan

[Installation]: RISC-V support?

1

### Your current environment ```text PyTorch version: N/A Is debug build: N/A CUDA used to build PyTorch: N/A ROCM used to build PyTorch: N/A OS: Fedora Linux 38 (Thirty Eight)...

SG2042 服务器 make 出错

1

/opt/llm/InferLLM/src/kern/optimized/rvv/kernel.cpp: Assembler messages: /opt/llm/InferLLM/src/kern/optimized/rvv/kernel.cpp:155: Error: unrecognized opcode `vsetivli x0,1,e32,m1', extension `v' or `zve64x' or `zve32x' required /opt/llm/InferLLM/src/kern/optimized/rvv/kernel.cpp:156: Error: unrecognized opcode `vmv.s.x v0,x0', extension `v' or `zve64x' or `zve32x' required /opt/llm/InferLLM/src/kern/optimized/rvv/kernel.cpp:162:...

Could TensorRT-LLM use only CPU for inference?

Could TensorRT-LLM run on RISC-V?

### 问题描述 Has anyone installed PaddleNLP successfully on RISC-V architecture?

[ 98%] Linking CXX executable ct2-translator /usr/bin/ld: ../libctranslate2.so.4.4.0: undefined reference to `cblas_sgemm' collect2: error: ld returned 1 exit status make[2]: *** [cli/CMakeFiles/translator.dir/build.make:98: cli/ct2-translator] Error 1 make[1]: *** [CMakeFiles/Makefile2:168: cli/CMakeFiles/translator.dir/all] Error...

support for qemu？

11

Does the project currently support interconnection of multiple qemu virtual machines? I installed exo in Qemu of multiple architectures, but they can't find each other.

Qemu VM infer error

1

Device: Qemu RISC-V virtual machine, with only CPU. I get an error when I execute the following command: DEBUG=9 exo run llama-3.1-8b error： **Error processing prompt: /tmp/tmp0a312mvx: undefined symbol: __gnu_f2h_ieee**...

Has anyone successfully installed TGI on a RISC-V architecture machine?

我按照文档尝试将 llama-2-7b 转换为 safetensors 模型或者 flm 格式，都出现以下错误： ValueError: Converting from Tiktoken failed, if a converter for SentencePiece is available, provide a model path with a SentencePiece tokenizer.model file.Currently available slow->fast...