🤝 Open to Collab

LH-Tech AI

LH-Tech-AI

74 26 657

https://lh-tech.de/ai/

LH-Tech-AI

AI & ML interests

Small AI and ML models. Trained by myself. Completely OpenSource. For you. | Reddit: https://www.reddit.com/user/LH-Tech_AI/

Recent Activity

reacted to theirpost with 🤯 about 2 hours ago

Supra2-100M is out! Go check it out: - https://www.reddit.com/r/LocalLLaMA/comments/1velyl9/new_models_supra2100m_base_and_instruct_go_check/ - https://huggingface.co/SupraLabs/Supra2-100M - https://huggingface.co/SupraLabs/Supra2-100M-Instruct Give us a like and a follow!! HAVE FUN 🤗🔥🚀 more coming soon...

reacted to theirpost with 🚀 about 2 hours ago

reacted to theirpost with 🔥 about 2 hours ago

View all activity

Organizations

reacted to their post with 🤯🚀🔥 about 2 hours ago

Post

Supra2-100M is out!
Go check it out:
- https://www.reddit.com/r/LocalLLaMA/comments/1velyl9/new_models_supra2100m_base_and_instruct_go_check/
- SupraLabs/Supra2-100M
- SupraLabs/Supra2-100M-Instruct

Give us a like and a follow!!

HAVE FUN 🤗🔥🚀
more coming soon...

posted an update about 2 hours ago

Post

replied to their post about 10 hours ago

Hey guys!
Supra2-100M base is delayed by another 4 hours. So it will be released at around 17PM in german time.
Why? Because we're running some tests, improving quality and tuning the model to deliver you the best quality possible! 🔥🤗
Stay tuned!
Instruction tuned model coming this evening (around 20PM in german time)! 😁

replied to their post about 10 hours ago

This comment has been hidden

replied to their post about 10 hours ago

i will benchmark later. i am doing the SFT data right now 🤗

replied to their post about 10 hours ago

lol. okay

reacted to their post with 🔥 about 10 hours ago

Post

2283

Announcing The Supra2 Family And Supra2-100M

Today, we are announcing a brand-new series of SupraLabs models: Supra2
This series will feature various models, including such as:
- 🐜 Supra2-Nano (0.4M) → The smallest Supra2 model.
- 🤏 Supra2-Small (1.4M) → The tiny model that runs everywhere.
- 💪 Supra2-Medium (25M) → Our medium class model in the Supra2 family. The powerful midsizer.
- 🔥 Supra2-Pro (100M): base, instruct, reasoning, code, math and more! → The most capable model yet! A real allrounder for all your everyday tasks.
- 🎨 Supra2-IMG → our generative text-to-image model
...and many more...

Current progress:
- Nano (0.4M) and Small (1.4M): in training; almost done. Baseline set.
- Medium (25M): coming soon...
- Pro (100M): in training; finishes in 66 hours - Monday, 3rd August 2026, 12:00AM
- IMG: coming soon...

You can support us with a like and follow if you want!
Don't miss our next release! Stay tuned...

26 replies

replied to their post about 10 hours ago

why do you think it's 111k?

replied to their post about 10 hours ago

13AM - sorry. lol... it took one hour longer. Base model coming at around 13:30AM. Instruct Model coming later this day. 🔥

replied to AxionLab-official's post about 14 hours ago

reacted to AxionLab-official's post with 🚀🔥 about 14 hours ago

Post

1170

Tommorow we are releasing Supra2-Pro(100M params), our flagship, get ready!

(follow SupraLabs to get tuned in!)

SupraLabs

9 replies

replied to their post 1 day ago

yessir!

reacted to AxionLab-official's post with 🧠👀 1 day ago

Post

11043

THIS IS CRAZY! THE MODEL ON THE IMAGE(Supra-50M-Reasoning) answered correctly and its QUANTIZED IN 2BIT! THE RESPONSE IS CORRECT, IN A 15MB SIZE FILE!

15 replies

replied to Banaxi-Tech's post 1 day ago

Yessir 🔥 🚀

replied to Banaxi-Tech's post 1 day ago

Thanks 🤗

replied to Banaxi-Tech's post 1 day ago

BananaMind Base Bench 1.1
Overall Elo: 992
Accuracy: 180/350 (51.43%)
Weighted accuracy: 46.89%
language_completion: Elo 1497 | 49/50 (98.00%) | weighted 98.73%
commonsense: Elo 989 | 30/50 (60.00%) | weighted 54.89%
world_knowledge: Elo 1003 | 31/50 (62.00%) | weighted 56.74%
context_tracking: Elo 904 | 19/50 (38.00%) | weighted 37.26%
quantitative: Elo 881 | 15/50 (30.00%) | weighted 28.08%
logical_reasoning: Elo 995 | 21/50 (42.00%) | weighted 36.05%
code_completion: Elo 959 | 15/50 (30.00%) | weighted 32.17%
Report: runs/bananamind-base-bench-1.1/report.json
Predictions: runs/bananamind-base-bench-1.1/predictions.md

LH-Tech AI

AI & ML interests

Recent Activity

Organizations

LH-Tech-AI's activity