Missing info/broken links
complete
A
Anonymous
There's a couple things missing.
"Learn more about this model" for sambanova-llama-3-405b-instruct leads to https://huggingface.co/neuralmagic/Meta-Llama-3-8B-Instruct-FP8. The other two have problems also. Corrected:
- meta-llama/Llama-3.1-405B-Instruct [link: https://huggingface.co/meta-llama/Llama-3.1-405B-Instruct]
- meta-llama/Llama-3.1-70B-Instruct [link: https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct]
- meta-llama/Llama-3.1-8B-Instruct [link: https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct]
Additionally,
- "llama3" should be Orenguteng/Llama-3.1-8B-Lexi-Uncensored-V2
- "qwen32-preview" should be Qwen/QwQ-32B-Preview
etc.
mistoria and eva-qwen are similar enough to be considered redundant. (Subjective)
Qwen-72b-instruct is notably missing. The finetunes, mistoria and eva-qwen are better at roleplay, but significantly worse in other tasks such as summarization and rewriting.
Finding the API key under clicking model name > python > api_key isn't intuitive, and should be improved.
Linkpharm
Canny AI
Merged in a post:
Model description
kitty
It'd be helpful to have a brief description of the model's intended use and recommended settings. It's not clear that the blue icon will take you to the HuggingFace repo.
M.R.
complete
M.R.
This is all amazing feedback and we are going to review it and break it down into individual task. Thank you for posting this!