Also, Colox is an open source experiment for reasoning under the MIT license. It has been fine tuned on LLaMA 3.1 8B Instruct.
NoName PRO
retronic
AI & ML interests
Creator of Colox and Platmoji! XD NOTE: All of my projects are under the MIT license, available for anyone to use as long as I am credited.
Organizations
None yet
replied to
their
post
5 months ago
posted
an
update
5 months ago
Post
255
I need your help. Colox is a VERY BIG MODEL, and unfortunately, even I don't have the processing power. If anyone is willing to run it for themselves and test it, please send me the performance. My email is [email protected] and the required VRAM is estimated to be 122GB. Thank you. :)
-
- 2 replies
posted
an
update
5 months ago
Post
192
UPDATE ABOUT COLOX:
I've intended to train Colox two times, the following will show a list of how many times I planned to train it: one with an open source database, then another with personality, safety filter, info about itself, and who created it. And plus, I am putting in a bit more info in the training so it knows a bit better. And I am happy to announce the first training is complete! I am making all models made from the two trainings open source, and the final training is the official release. NOTE: I HAVE NOT TESTED THIS YET, THIS MODEL MAY NOT REASON AND IF THAT IS THE CASE, I WILL FIX IT.
I've intended to train Colox two times, the following will show a list of how many times I planned to train it: one with an open source database, then another with personality, safety filter, info about itself, and who created it. And plus, I am putting in a bit more info in the training so it knows a bit better. And I am happy to announce the first training is complete! I am making all models made from the two trainings open source, and the final training is the official release. NOTE: I HAVE NOT TESTED THIS YET, THIS MODEL MAY NOT REASON AND IF THAT IS THE CASE, I WILL FIX IT.
posted
an
update
6 months ago
Post
308
Hello, I know it has been a bit since my last post. I've realized my Open Research project is very big so I will stop production. But never say never. I am also starting to try Colox again. The wait for updates may be long as I am busy working on an AI learning tool called Brixallent, but we don't need to focus on that right now. :)
posted
an
update
10 months ago
Post
1837
The Colox idea is getting replaced with a clone of OpenAI Deep Research due to retraining issues and reasoning issues
So now I am working on a Deep Research system with Ollama that will function like OpenAI's version for FREE! This will be a local alternative, no potato PC can handle this keep in mind.
So now I am working on a Deep Research system with Ollama that will function like OpenAI's version for FREE! This will be a local alternative, no potato PC can handle this keep in mind.
posted
an
update
10 months ago
Post
3040
Colox will be retrained due to performance tests not meeting requirements
replied to
their
post
10 months ago
i really did tests, it is not smarter due to the way i trained it. i am working on it a lot as it did not meet my expectations.
replied to
their
post
10 months ago
colox will be delayed due to bugs and issues
replied to
their
post
10 months ago
idk why though, my mind decided to go and use llama 2.
posted
an
update
10 months ago
Post
1982
Colox is out, may be bugs!
Colox is out and ready on HF, it might have bugs though as it is not tested yet. You can try for yourself now! :)
Colox is out and ready on HF, it might have bugs though as it is not tested yet. You can try for yourself now! :)
reacted to
their
post with 😎
10 months ago
Post
4617
Colox, a reasoning AI model. I am currently working on a model smarter than GPT o1 that thinks before it speaks. It is coming tomorrow in the afternoon.
-
- 7 replies
replied to
their
post
10 months ago
note: benchmarks will be out on like tuesday
reacted to
their
post with 🔥
10 months ago
Post
4617
Colox, a reasoning AI model. I am currently working on a model smarter than GPT o1 that thinks before it speaks. It is coming tomorrow in the afternoon.
-
-
-
- 7 replies