Hugging Face's logo

Orion LLM Labs
community
Activity Feed

AI & ML interests

Advancing AGI through efficient local LLM inference.

Recent Activity
DedeProGames  updated a model 15 days ago
OrionLLM/GRM-3.2-Turf
DedeProGames  updated a model 15 days ago
OrionLLM/GRM-3.2-Cliff
DedeProGames  updated a model 15 days ago
OrionLLM/GRM-3.2-Sky
View all activity
Banaxi-Tech 
posted an update about 11 hours ago
view post
Post
50
We're introducing ACR 1.0.
We trained this model on a 5070 Ti for weeks, here are some of the architecture details:
57M parameters, with one M and one G stream.
When we tested it on benchmarks, we got these results:
Benchmark Full G-Only Delta
PIQA 62.24% 53.43% +8.81
ARC-Easy 41.96% 32.28% +9.68
HellaSwag 33.19% 29.08% +4.11
Tiny ToM 40.65% 33.75% +6.90
ArithMark 3.0 33.40% 32.80% +0.60
Base Bench 1.1 51.71% 40.29% +11.42

Check it out at saicr/ACR-1.0
  • 3 replies
·
DedeProGames 
posted an update 1 day ago
view post
Post
2669
Im working on a 23M ASR model, trained on 100k hours of audio
  • 4 replies
·
Banaxi-Tech 
posted an update 1 day ago
view post
Post
155
Its SAICR time tomorrow.
Get ready!
saicr
  • 28 replies
·
Banaxi-Tech 
posted an update 2 days ago
view post
Post
3035
We will release the BEST SLM Leaderboard before Oct 11.

It will feature everything:
Easy to use model picker.
EXTREMELY Easy way to add your own models (2 click)
MULTIPLE leaderboard for different model types

And more!


So why don't you help us build it?

Join
betu-slm-leaderboard-testers
  • 3 replies
·
DedeProGames 
posted an update 5 days ago
view post
Post
5814
how is this possible
  • 13 replies
·
Banaxi-Tech 
posted an update 6 days ago
view post
Post
5153
ACR 1.0 launch is being prepared and researched now!
Also I'm going to vacation tomorrow but it should still be released!

saicr
  • 3 replies
·
DedeProGames 
posted an update 8 days ago
view post
Post
6279
🧱 SLM Tetris Arena: can a small language model play Tetris without ever being trained on it?

I built an arena where tiny decoder-only LMs (50K–250M params) play Tetris zero-shot. There is no fine-tuning and no game data. They only use what they picked up from pre-training on text.

How it works:
- For every piece, the engine simulates each legal placement and describes the result in plain English ("clears one line, creates no new holes, keeps the stack low…").
- The model never sees the grid. It reads each description, and the arena compares log P(" good move") with log P(" bad move"). The best-rated placement is played.
- Every player gets the same piece sequence, so it's a fair race.
- There are two protocols: Guided (the rules are in the prompt) and Blind (no rules, only pre-training knowledge).

Two ways to play:
- Match: pick any models (even your own, custom architectures welcome) and watch them play side by side on retro 8-bit boards.
- Ranked: press Play and the arena picks up to 4 models at random from a curated pool of 29. Nobody chooses their opponents, so Elo can't be farmed. Matches run on the server and count even if you close the tab.

First results (~225 ranked matches):
- gpt2 (124M) leads with 1283 Elo, but SupraNeo-4M (4M) is right behind at 1239. Next come LowOnMind-5M and BananaMind-2.1-Pico (1.5M!).
- Model size barely predicts Elo (r ≈ 0.06). Survival does (r ≈ 0.9): the models that avoid holes and keep the stack low are the ones that win.

Every ranked match (seed, model commit SHAs, scores, Elo before/after) is logged in a public dataset.

▶ Play: DedeProGames/SLM-Tetris-Arena
📊 Results: DedeProGames/lm-tetris-arena-results

Want your model in the Ranked pool? Drop it in the comments!
  • 1 reply
·
Banaxi-Tech 
posted an update 10 days ago
view post
Post
7607
We're releasing a MAJOR update to the BananaAll SLM Super App.
If you want to use a custom architecture, previously you had to go trough reviewing the code yourself, now add an Openrouter API key and review it with GPT 6 Luna in one button. A review cost be half a cent so anyone can try it. This is one of the main features.
Now ROCm, AMD and Windows, Mac support.
Colab and Molab support.

Detailed list of features:
Get improved Windows Python detection and support paths for compatible AMD ROCm, Intel XPU, and Apple MPS setups.
Choose local training or export a self-contained Python script for Colab or Molab. Notebook runs produce a downloadable model ZIP.
Start pretraining with an existing model’s tokenizer, or train a new one from your datasets.
Try experimental 1.58-bit Ternary fake-quantized training on NVIDIA GPUs.
Watch live tokens per second. Model compilation is on by default and falls back automatically if it fails.
Build custom architectures with separate configuration and modeling files, then review the training code manually or with optional OpenRouter AI Review.
Install from source with the new coding-agent instructions.
This release also fixes inflated loss reporting for custom models.



And for those users who didn't want to try it out just because installation would be so hard, it isnt now.
Go to any coding agent (Pi, Claude Code, Codex, OpenCode, basically all work), and just paste "Install BananaAll for me. Fetch and follow https://raw.githubusercontent.com/BananaMind/BananaAll/main/agent_install.txt."
That's it.

Check it out at https://github.com/BananaMind/BananaAll/

Also on SAICR, we're currently training a new major model (NACR v2) and ACR 1.0 is in the finishing.

  • 11 replies
·
DedeProGames 
posted an update 10 days ago
view post
Post
2882
🧱 SLM Tetris Arena: can a small language model play Tetris without ever being trained on it?

I built an arena where tiny decoder-only LMs (50K–250M params) play Tetris zero-shot. There is no fine-tuning and no game data. They only use what they picked up from pre-training on text.

How it works:
- For every piece, the engine simulates each legal placement and describes the result in plain English ("clears one line, creates no new holes, keeps the stack low…").
- The model never sees the grid. It reads each description, and the arena compares log P(" good move") with log P(" bad move"). The best-rated placement is played.
- Every player gets the same piece sequence, so it's a fair race.
- There are two protocols: Guided (the rules are in the prompt) and Blind (no rules, only pre-training knowledge).

Two ways to play:
- Match: pick any models (even your own, custom architectures welcome) and watch them play side by side on retro 8-bit boards.
- Ranked: press Play and the arena picks up to 4 models at random from a curated pool of 29. Nobody chooses their opponents, so Elo can't be farmed. Matches run on the server and count even if you close the tab.

First results (~225 ranked matches):
- gpt2 (124M) leads with 1283 Elo, but SupraNeo-4M (4M) is right behind at 1239. Next come LowOnMind-5M and BananaMind-2.1-Pico (1.5M!).
- Model size barely predicts Elo (r ≈ 0.06). Survival does (r ≈ 0.9): the models that avoid holes and keep the stack low are the ones that win.

Every ranked match (seed, model commit SHAs, scores, Elo before/after) is logged in a public dataset.

▶ Play: DedeProGames/SLM-Tetris-Arena
📊 Results: DedeProGames/lm-tetris-arena-results

Want your model in the Ranked pool? Drop it in the comments!
  • 1 reply
·
DedeProGames 
posted an update 12 days ago
view post
Post
151
🚀 Training GPT-U-20M on 2.6B tokens
Banaxi-Tech 
posted an update 12 days ago
view post
Post
4867
We're excited to release BananaAll, our SLM Super App.

It allows you to do EVERYTHING you need to do to trains SLMs in a single app, no terminal, no 30 chrome tabs.

The train tab allows you to train models, select datasets from presets, and use other ones with auto mapping, model size slider, it automatically generates a training script for you.

Then after you've trained the model or want to compare it to competitors, the evaluation tab, run ARC EASY, ARC Challenge, Hellaswag, PIQA, Arithmark 3, BananaMind Base Bench and more! Simple Results screen.

And lastly the inference tab, run your trained models or others.

Normally you would need seperate apps or scripts for that, but the BananaAll Super App lets you do all of that in a single app.

We also trained a small 2.5M parameter model on 200M tokens of Fineweb edu, The results: BananaMind Base Bench 854 and 53% on PIQA. On only 200M tokens.

Check it out at https://github.com/BananaMind/BananaAll.
  • 25 replies
·
DedeProGames 
posted an update 13 days ago
view post
Post
96
🚀 Possible new model in the GRM-3.2 family — GRM-3.2-Mist

A new model may be joining the GRM-3.2 family, originating from an experimental finetune currently under evaluation.

If this model passes internal testing and demonstrates strong performance, it will be released as GRM-3.2-Mist — a portable 1.7B model targeting reasoning and agentic coding tasks.

The model already exists and is currently undergoing evaluation. Follow this page for further updates:

OrionLLM
Banaxi-Tech 
posted an update 13 days ago
view post
Post
1900
hi everyone
we have released nacr
its not just any model, its nacr
we have 6 more features and this model only uses 20% of its total capacity!
check it out at saicr/nacr
we're currently working on expanding access as we do more research but right now you have to use our gated access form


follow
saicr if you're interested
if you want to join saicr, first read the entire nacr readme, then press the join button.
  • 14 replies
·
Banaxi-Tech 
posted an update 14 days ago
view post
Post
1530
saicr is going to have its first model launch around October 2.
We're working so hard to get the models available as soon as possible.
  • 7 replies
·
DedeProGames 
updated 3 models 15 days ago
Banaxi-Tech 
posted an update 15 days ago
view post
Post
102
This day is Sol nice.
DedeProGames 
in OrionLLM/GRM-3.2-Sky 16 days ago

This is a good model, This is my assessment of the model

5
#8 opened about 1 month ago by
funnygeeker
Banaxi-Tech 
posted an update 16 days ago
view post
Post
104
We have some updates to @BananaMindBot 🍌
It can now train models, ask it to train a model, and i will train it for you.
It now can also merge PRs And like models.
  • 28 replies
·