WasmEdge / WasmEdge

WasmEdge is a lightweight, high-performance, and extensible WebAssembly runtime for cloud native, edge, and decentralized applications. It powers serverless apps, embedded functions, microservices, smart contracts, and IoT devices.

Home Page:https://WasmEdge.org

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

bug: Failed to run `openhermes-2.5-mistral-7b` with the `wasi-nn` plugin of `0.13.5`

apepkuss opened this issue · comments

Summary

The latest wasi-nn plugin failed to run openhermes-2.5-mistral-7b. And the behaviors on macOS-14.2 (m1 pro) and Ubuntu-22.04 (Intel(R) Xeon(R) Platinum 8259CL CPU), respectively, are different:

  • On macOS
    image

  • On Ubuntu
    image

Current State

No response

Expected State

No response

Reproduction steps

  • Install WasmEdge 0.13.5 + ggml plugin of latest version
  • Download openhermes-2.5-mistral-7b
    curl -LO https://huggingface.co/second-state/OpenHermes-2.5-Mistral-7B-GGUF/resolve/main/openhermes-2.5-mistral-7b.Q5_K_M.gguf
  • Download llama-chat.wasm
    curl -LO https://github.com/second-state/llama-utils/raw/dev/chat/llama-chat.wasm
  • Run the model
    wasmedge --dir .:. --nn-preload default:GGML:AUTO:openhermes-2.5-mistral-7b.Q5_K_M.gguf llama-chat.wasm -p chatml -r '<|im_end|>'

Screenshots

DESCRIPTION

Any logs you want to share for showing the specific issue

No response

Components

Others

WasmEdge Version or Commit you used

0.13.5

Operating system information

Ubuntu 22.04, macOS 14.2

Hardware Architecture

x86_64, aarch64

Compiler flags and options

No response