Spaces:
Running
[MODELS] Discussion
Here we can discuss about HuggingChat available models.
what are limits of using these? how many api calls can i send them per month?
How can I know which model am using
How can I know which model am using
at the bottom of your screen:
Out of all these models, Gemma, which was recently released, has the newest information about .NET. However, I don't know which one has the most accurate answers regarding coding
Gemma seems really biased. With web search on, it says that it doesn't have access to recent information asking it almost anything about recent events. But when I ask it about recent events with Google, I get responses with the recent events.
apparently gemma cannot code?
Gemma is just like Google's Gemini series models, it have a very strong moral limit put on, any operation that may related to file operation, access that might be deep, would be censored and refused to reply.
So even there are solution for such things in its training data, it will just be filtered and ignored.
But still didn't test the coding accuracy that doesn't related to these kind of "dangerous" operations
It seems that UI improvements are being considered.
https://huggingface.co/spaces/huggingchat/chat-ui/discussions/682
can you replace llama 3.3 with llama 4 or can you add qvq or qwen 2.5vl ?
can you do better interface in all site i mean! with dashbord! its very hard to concertrate on different models in main menu! and searching models is difficult now! also can you fix DeepSeek-R1-32B often generate strange, unreadable output when responding in Japanese — with mixed languages or garbled characters. This suggests limited multilingual support. At least for Japanese, their performance is far behind models like Gemma or Phi, and in some languages they become nearly unusable.
also can you fix DeepSeek-R1-32B often generate strange, unreadable output when responding in Japanese
This may be fixed by adjusting the system prompt. Adding something like “Respond in Japanese” will result in many models responding in Japanese.
For now, I think it's difficult to know where to raise requests, so I'll organize them based on what I know.
For general issues such as inconveniences or QoL when using Hugging Face GUI (Excluding library-related issues)
https://github.com/huggingface/hub-docs/issues
https://huggingface.co/posts/victor/964839563451127
Hugging Face Discord
[email protected]
For libraries issues
https://github.com/orgs/huggingface/repositories
https://github.com/gradio-app/gradio/issues
https://github.com/pytorch/pytorch/issues (Not by Hugging Face but important)
For Serverless Inference API
For Zero GPU
https://huggingface.co/spaces/zero-gpu-explorers/README/discussions
For Hugging Chat
https://huggingface.co/spaces/huggingchat/chat-ui/discussions (here)
https://github.com/huggingface/chat-ui/issues
For payment / billing issues
And so on...
MY ASSISTANT DOESNT ANSWER CORRECTLY!
that doesnt work! i change it but it thinks and answers in japanese and chinese and something not logic!
THIS IS MY SETTINGS:You are Qwen, created by Alibaba Cloud. You are a helpful assistant. think/reson and answer in russian languange! IF I AM NOT CORRECTLY DO SOMETHING WRITE ME!
Hmm... He answered in Russian...
guys, we need to help mr. sarrazin out. he's all alone in this project. we need new contributors to submit PRs to the HuggingChat official github repository.
in order to push forward to adding new features and fixing bugs much sooner.
MY ASSISTANT DOESNT ANSWER CORRECTLY!
that doesnt work! i change it but it thinks and answers in japanese and chinese and something not logic!
THIS IS MY SETTINGS:You are Qwen, created by Alibaba Cloud. You are a helpful assistant. think/reson and answer in russian languange! IF I AM NOT CORRECTLY DO SOMETHING WRITE ME!
This is not only seen with the models - “Qwen”; perhaps it is a “learning cost” of the models.