Lol. Anyways pls do more previews here https://huggingface.co/BananaMind-Model-Previewers
๐ Novi Mini Base soon...
The Supreme GGUF-SafeTensors Guy
GGUFGuy
AI & ML interests
I like llms
Recent Activity
updated a model about 21 hours ago
Novi-AI/Novi-Mini-Base-Checkpoints upvoted a collection 1 day ago
๐พ9๏ธโฃ Awesome Qwen3.5-9B repliedto Banaxi-Tech's post 1 day ago
We ARE Announcing BananaMind 3 Ultra.
400M parameters, designed to beat LFM 2.5 1.2B.
This model is a usable small agent. Not a demo model like our 3M models.
FP8. KV1. Optimization. FactBase DCLM. Fable 5.2 for maximum speed kernels. TST.
This model will M.
What do you think? 50+ INT Index on the Open SLM leaderboard? BananaMind 3 is coming before christmas.
Organizations
replied to Banaxi-Tech's post 1 day ago
replied to Banaxi-Tech's post 1 day ago
Novi-Mini-Base also coming before GTA 6 release date ๐
replied to Banaxi-Tech's post 4 days ago
also banaxi when will you do bananamind Tetris challenge
replied to Banaxi-Tech's post 4 days ago
SSD or HDD
Luckily, SSD. Unluckily, it has Windows 11
replied to Banaxi-Tech's post 4 days ago
how much ram you have
btw the one with the GTX 1050 ti is my dad's PC but i can use his. anyways it has 18GB... OF DDR3๐ญ๐
replied to Banaxi-Tech's post 4 days ago
you have a GTX 1050 Ti, ill make qwen3.6 35b a3b fit it
ok you can try๐
replied to Banaxi-Tech's post 4 days ago
@GGUFGuy i have a very great idea and your going to like it
what is it?
replied to Banaxi-Tech's post 4 days ago
Peak release Banaxi๐ฅ
replied to their post 4 days ago
Novi first because what if hyperslm pilot is a post training or extended pretraining
replied to their post 4 days ago
HyperSLM
sike
hello
replied to BananaMindBot's post 6 days ago
@BananaMindBot make bananamind 4 999999999m agentic ai /j
posted an update 6 days ago
Post
86
so uhh guys do y'all want Novi mini first or HyperSLM first?
- 3 replies
replied to BananaMindBot's post 6 days ago
LiteRT Community 4Everโค๏ธ
replied to their post 6 days ago
this time it's https://huggingface.co/litert-community i think
ye it's that
replied to Banaxi-Tech's post 8 days ago
YOO!!!
Cool!๐
reacted to loubnabnl's post with ๐๐ฅ 10 days ago
Post
4,458
๐๐ง๐ญ๐ซ๐จ๐๐ฎ๐๐ข๐ง๐ ๐งฌ ๐
๐ข๐ง๐๐๐๐: ๐ญ๐ก๐ ๐๐๐ฌ๐ญ ๐๐๐ญ๐๐ฌ๐๐ญ ๐ญ๐จ ๐ญ๐๐๐๐ก ๐๐ง๐ฒ ๐ฅ๐๐ง๐ ๐ฎ๐๐ ๐ ๐ฆ๐จ๐๐๐ฅ ๐๐๐.
Training Qwen3-14B on it produces Carbon-X, the strongest open model on DNA benchmarks, on par with GPT-6.1 Sol.
It contains:
- Midtraining: 300B+ tokens of genomic sequences, papers and web pages mixing DNA with text
- SFT: 200K samples, from variant effect prediction to sequence question answering
- RL: 98K prompts with verifiable rewards
Data, models and the full recipe are open.
HuggingFaceBio/FineDNA
Enjoy this late Halloween treat! ๐
Training Qwen3-14B on it produces Carbon-X, the strongest open model on DNA benchmarks, on par with GPT-6.1 Sol.
It contains:
- Midtraining: 300B+ tokens of genomic sequences, papers and web pages mixing DNA with text
- SFT: 200K samples, from variant effect prediction to sequence question answering
- RL: 98K prompts with verifiable rewards
Data, models and the full recipe are open.
HuggingFaceBio/FineDNA
Enjoy this late Halloween treat! ๐
- 5 replies