okay chatgpt
Mithula Artigala PRO
mithulaartigala
AI & ML interests
I love RL!
Recent Activity
new activity 1 day ago
speridlabs/idol:RTX 5070 SPEED new activity 1 day ago
speridlabs/idol:Add RTX 5070 community benchmark (41.8 s median) upvoted a paper 4 days ago
gpt-oss-120b & gpt-oss-20b Model CardOrganizations
None yet
replied to their post 5 days ago
replied to their post 5 days ago
the cloud one's have about 1m.
and they can compact and store it as well.
replied to their post 6 days ago
smarrrrtt. I immediately forget what my teachers say. cant blame qwen.
replied to their post 7 days ago
thats kinda kinda crazy coming out of you when YOUR NAME IS LITERALLY GGUFGUY😭
posted an update 7 days ago
Post
3,095
here is my qwen is BETTER than the human brain:
human-brain-bf16.gguf:
- no cuda support
- 3.8B old model BTW
- BAD information retention
- the gguf files doesn't even load
- no Q4KM (god did this intentionally to piss off the llama.cpp community trust)
- no finetunability :(
- NOT opensource
- training code is hidden, AND expects us to reverse engineer it (genuinely messed up, and disrespectful to the opensource community)
now, qwen on the other hand:
- available cuda support ✅
- very very new (and gets contant updates) ✅
- VERY good information retention ✅
- GGUF files loads properly ✅
- Q4_K_M support available ✅
- very fine tunable ✅
- open source as it can possibly get ✅
- many code and datasets are free and open source ✅
verdict: qwen beter ✅✅✅ 👍
human-brain-bf16.gguf:
- no cuda support
- 3.8B old model BTW
- BAD information retention
- the gguf files doesn't even load
- no Q4KM (god did this intentionally to piss off the llama.cpp community trust)
- no finetunability :(
- NOT opensource
- training code is hidden, AND expects us to reverse engineer it (genuinely messed up, and disrespectful to the opensource community)
now, qwen on the other hand:
- available cuda support ✅
- very very new (and gets contant updates) ✅
- VERY good information retention ✅
- GGUF files loads properly ✅
- Q4_K_M support available ✅
- very fine tunable ✅
- open source as it can possibly get ✅
- many code and datasets are free and open source ✅
verdict: qwen beter ✅✅✅ 👍
- 11 replies
replied to Banaxi-Tech's post 10 days ago
nothing. forget about it.
replied to Banaxi-Tech's post 10 days ago
HF cannot be real bro 😭
replied to Banaxi-Tech's post 10 days ago
SON 😭😭😭😭
replied to Banaxi-Tech's post 10 days ago
ohh yeah. gemini models should be completely discontinued and introduced with a new family.
There is NO coming back for gemini lol
replied to Banaxi-Tech's post 10 days ago
would be cool to see a good 90M model though ooohh
replied to Banaxi-Tech's post 10 days ago
so you are telling me his models are better than the gemma4 line up?
is it going to be the "gpt-x3 killer"?😂
alr alr im sorry
replied to Banaxi-Tech's post 10 days ago
ohhh no. dont kill gpt-x3. we love it.
replied to Banaxi-Tech's post 10 days ago
omg frfr
reacted to Banaxi-Tech's post with 🧠👍 10 days ago
Post
4,895
BananaMind is in the google situation before argon right now, right?
- 29 replies
posted an update 11 days ago
Post
196
I released mithulaartigala/calibrated-decisions-v0.1-20k
I am sorry that i couldn't reach the 50K goal, but I believe 20K samples are enough to get started.
Lets get the encoders training!
I am sorry that i couldn't reach the 50K goal, but I believe 20K samples are enough to get started.
Lets get the encoders training!
replied to their post 11 days ago
FYI (for the ppl who take out of context screenies): that stuff right there is bainfuck code 😭
replied to their post 11 days ago
my brain's fucked
posted an update 12 days ago
Post
3,272
We need to make software where people can start using decision models.