Maincode

/

Maincoder-1B

like 91

Follow

Maincode 47

say not supported by llama.cpp when I tried to convert it

WARNING:hf-to-gguf:Trying to load config.json instead
INFO:hf-to-gguf:Model architecture: MaincoderForCausalLM
ERROR:hf-to-gguf:Model MaincoderForCausalLM is not supported

and https://huggingface.co/spaces/ggml-org/gguf-my-repo errored out

I know, i tried that too. I am guessing that you need to modify llama.cpp to be compatible with MaincoderForCausalLM

Someone from our team is already working on it so there should be an update soon : )

Okay! Hope everything goes well!

Any update on this?

Hi.. We've been taking a break after the release so this could some time but rest assured it will up updated here the moment it is done and released....

But again, in the meantime, to encourage community involvement and participation too, I think it would be great to see someone from the community do that if they're up for the task... The model's license permits that too so you're all cool...

https://huggingface.co/Maincode/Maincoder-1B-ONNX

That PR has been merged. It's supported now! https://github.com/ggml-org/llama.cpp/pull/18534

https://huggingface.co/yourGGUF/Maincoder-1B_GGUF

https://huggingface.co/Maincode/Maincoder-1B-GGUF

Neat, now I can run the model locally :)

so i tried to load the model on lm studio (thinking that it would work) and i got this

🥲 Failed to load the model

Failed to load model

error loading model: error loading model architecture: unknown model architecture: 'maincoder'

Maybe enable LM Studio's beta channel. You need llama.cpp version >= b7614

https://github.com/ggml-org/llama.cpp/releases/tag/b7614

I set LM Studio to beta, but it still doesn’t work, and I’m not sure how to set up llama.cpp b7614 to work with LM Studio.

Probably need to wait a few days to a week for LM Studio to catch up if you don't want to run llama.cpp directly

okay