2 points | by openkame 8 hours ago ago
1 comments
Good luck with model compatibility, the inevitable fracturing of the project when llama, qwen, and gpt are just not possible.
You probably need >1000 people to maintain this project.
What's the value add over what ollama already does?
We're also going containerization because "hacking" is a thing.
Good luck with model compatibility, the inevitable fracturing of the project when llama, qwen, and gpt are just not possible.
You probably need >1000 people to maintain this project.
What's the value add over what ollama already does?
We're also going containerization because "hacking" is a thing.