-
currently: the model downloads at instantiation.
target: the model downloads when generation is first required
advantages:
1. removes a delay in discovering invalid configuration
2. removes de…
-
### Feature Name
Research about Stability.ai
### Feature Description
This a research about Stability.ai, learning more about its supported models, how it is used and many more
### Motivati…
-
Some models handle the task generation and checking much more effectively than others. It would be useful to be able to switch API connections (or models with open router) to pick one specifically fo…
-
Hi openai devs,
Congrats on releasing o1 models, and I'm excited to test them out. Since there seems to be new "reasoning" tokens in generation, I wonder if there is a new version for tiktoken for …
-
This is an amazing work. I have been working on something that would require me to evaluate the generated outputs of models like Mistral, using a prompt like:
`"Fill the [MASK] token in the sentence.…
-
### Model description
I am not sure these new models from OpenGVLab are supported here.
[OpenGVLab/InternVL2-8B](https://huggingface.co/collections/OpenGVLab/internvl-20-667d3961ab5eb12c7ed1463e)
…
-
Forge recently changed how models are loaded, which was a sensical thing to do as selecting a flux model will take forever to load in the background. My idea is simple.
If it’s possible, make it s…
-
**Project description**
Nexa SDK is a comprehensive toolkit for supporting ONNX and GGML models. It supports text generation, image generation, vision-language models (VLM), auto-speech-recognition…
-
Dear Author's,
I am currently working on a project focused on molecular generation using diffusion models. I have come across the code for the two-stage generation process from your research on Gene…
-
Error appears while running generation. I tried to research this error and found references about other models file not downloading correctly. But I made sure that they are and downloaded more than on…