Bin-Huang / chatbox

User-friendly Desktop Client App for AI Models/LLMs (GPT, Claude, Gemini, Ollama...)
https://chatboxai.app
GNU General Public License v3.0
20.7k stars 2.1k forks source link

[Feature] Read one or multiple files #1214

Open trymeouteh opened 6 months ago

trymeouteh commented 6 months ago

Problem Description Unable to add files to the prompt

Proposed Solution A button that can add one or more files to the prompt which can be submitted with the prompt

Additional Context When using Ollama models, there is no option to send files and there are models that support this in Ollama such as llava

mahomedh commented 6 months ago

I was also looking for a way to add attachments to the prompt. This was a useful feature when using the ChatGPT interface.

mlepeshkin commented 2 months ago

File attachment feature for OpenAI models is super needed.

mahomedh commented 2 months ago

This feature has been available since version 1.3.4 https://chatboxai.app/help-center/changelog so I think this issue can be closed now?

mlepeshkin commented 2 months ago

For me it's about OpenAI's API. And in cuvvent v. 1.3.12 it's not implemented. When I'm trying to attach files I'm getting: "The current model gpt-4o does not support sending files. Currently supported models: Chatbox AI 4."

dmanresa-saes commented 2 months ago

I am experiencing the same issue. I have tried with versions 1.3.4, 1.3.10, and 1.3.14 and I always get the same message. I tested version 1.3.4 with GPT-4 because GPT-4o did not support it yet, but it gives the same error.