If they give some kind of GPU time for training, I would be so happy (although it would probably be pro only)
CodeSoft PRO
CodeSoft
AI & ML interests
Working on small language models
Recent Activity
updated a bucket about 7 hours ago
CodeSoft/slm-arena-data liked a model 5 days ago
internlm/Intern-S2-397B upvoted an article 6 days ago
How Did We Get Here?Organizations
reacted to SeaWolf-AI's post with π₯ 16 days ago
Post
4065
Introducing the Global LLM Download Leaderboard π
Cumulative download counts are a museum. They reward age, not relevance β a model released two years ago can sit near the top on the strength of downloads it earned long before anyone stopped using it. If you want to know what the open LLM ecosystem is actually running today, you need a different lens.
So we built one. The Global LLM Download Leaderboard ranks text-generation models by their trailing 30-day downloads, measured directly from the Hugging Face API and refreshed every day.
π VIDraft/global-llm-leaderboard
Why a 30-day window changes what you see
A cumulative chart answers "what has been popular." A 30-day chart answers "what is being adopted right now." Those are very different questions β and the second one is the one that matters if you're deciding what to build on, quantize, fine-tune, or serve this quarter. Momentum, not history.
What it shows
Global Top 300, with tabs for πΊπΈ USA Β· π¨π³ China Β· πͺπΊ EU
Six share-of-download charts: by country, by parameter size, by quantization, by type (Base / Instruct / Quantized / MoE), by release year, and by organization (Top 10)
Per-model chips for parameter size, quantization, license, and type
English / νκ΅μ΄ with automatic browser-language detection and a manual toggle
What the data reveals
The frontier is bipolar. Two countries account for the large majority of the top-300's 30-day downloads. Open-model gravity is concentrating, not dispersing.
Small is winning. A striking share of all downloads goes to sub-3B models β the clearest signal yet that on-device and cost-efficient deployment, not maximum parameter count, is driving real-world adoption.
Quantization is mainstream. GGUF, AWQ, FP8 and friends aren't a niche β a large fraction of the most-downloaded artifacts are quantized, because that's what people actually run.
Benchmarks measure what a model can do. Downloads measure what people choose to use.
Cumulative download counts are a museum. They reward age, not relevance β a model released two years ago can sit near the top on the strength of downloads it earned long before anyone stopped using it. If you want to know what the open LLM ecosystem is actually running today, you need a different lens.
So we built one. The Global LLM Download Leaderboard ranks text-generation models by their trailing 30-day downloads, measured directly from the Hugging Face API and refreshed every day.
π VIDraft/global-llm-leaderboard
Why a 30-day window changes what you see
A cumulative chart answers "what has been popular." A 30-day chart answers "what is being adopted right now." Those are very different questions β and the second one is the one that matters if you're deciding what to build on, quantize, fine-tune, or serve this quarter. Momentum, not history.
What it shows
Global Top 300, with tabs for πΊπΈ USA Β· π¨π³ China Β· πͺπΊ EU
Six share-of-download charts: by country, by parameter size, by quantization, by type (Base / Instruct / Quantized / MoE), by release year, and by organization (Top 10)
Per-model chips for parameter size, quantization, license, and type
English / νκ΅μ΄ with automatic browser-language detection and a manual toggle
What the data reveals
The frontier is bipolar. Two countries account for the large majority of the top-300's 30-day downloads. Open-model gravity is concentrating, not dispersing.
Small is winning. A striking share of all downloads goes to sub-3B models β the clearest signal yet that on-device and cost-efficient deployment, not maximum parameter count, is driving real-world adoption.
Quantization is mainstream. GGUF, AWQ, FP8 and friends aren't a niche β a large fraction of the most-downloaded artifacts are quantized, because that's what people actually run.
Benchmarks measure what a model can do. Downloads measure what people choose to use.
- 5 replies
reacted to KlondikeDev's post with π₯ 16 days ago
Post
2412
Boris-1.7-D60M-n30M out NOW!
opencerebral/Boris-1.7-D60M-n30M
Following this will be Boris-1.8-D60M-n30M, which will test both the n-gram embeddings AND a new architecture.
Then, Boris-2 will begin training!
opencerebral/Boris-1.7-D60M-n30M
Following this will be Boris-1.8-D60M-n30M, which will test both the n-gram embeddings AND a new architecture.
Then, Boris-2 will begin training!
replied to their post 16 days ago
Would you allow me to duplicate your SLM Arena space if i make changes to it? I see its under MIT but dont wanna end up like Multivex if you didnt mean to set it to MIT?
No, I intended for it to be MIT. You can absolutely duplicate it, just please have a visible attribution.
replied to their post 16 days ago
Accepted
replied to their post 16 days ago
Okay, but the invite says it expired. Could you please resend it?
replied to their post 16 days ago
Thanks for reaching out! Since I'm not part of the BananaMind project, could you clarify what kind of feedback you're looking for?
posted an update 17 days ago
Post
2666
Sorbet Mini Experimental is out!
It's not the best model in the 5M parameter range, but it's going to be a really useful model to train on top of. This release was mostly to prove that the model actually works. I trained it on 150M tokens from TinyStories in about 12 minutes.
Does anyone have any tips on training models in this size range? I want to make the full Sorbet Mini release as good as it can be.
It's not the best model in the 5M parameter range, but it's going to be a really useful model to train on top of. This release was mostly to prove that the model actually works. I trained it on 150M tokens from TinyStories in about 12 minutes.
Does anyone have any tips on training models in this size range? I want to make the full Sorbet Mini release as good as it can be.
- 13 replies
reacted to Banaxi-Tech's post with π₯ 17 days ago
Post
2894
We're excited to release BananaMind 2.1 Pico Preview!
It includes the first preview of our BananaMind 2.1 architecture!
This model gets near BananaMind 2 Micro performance at half the parameters and 37.5x less tokens!
Thats insane!
The current architectures includes about 500K parameters of the total 1.5M parameters in n-gram embeddings and the layer 2 is run twice.
It also includes XSA and the XSA refresh gate.
We're still going to improve the architecture in the final release.
Check it out at:
Follow us for more models:
BananaMind
@Banaxi-Tech
@vovaRL
@DedeProGames
It includes the first preview of our BananaMind 2.1 architecture!
This model gets near BananaMind 2 Micro performance at half the parameters and 37.5x less tokens!
Thats insane!
The current architectures includes about 500K parameters of the total 1.5M parameters in n-gram embeddings and the layer 2 is run twice.
It also includes XSA and the XSA refresh gate.
We're still going to improve the architecture in the final release.
Check it out at:
Follow us for more models:
@Banaxi-Tech
@vovaRL
@DedeProGames
- 3 replies
reacted to AtAndDev's post with π₯ 17 days ago
Post
2826
SPECK 2 IS ALREADY OUT: specklabs/Speck2-140M
Pretrained on 4x more tokens than the previous releases (20b vs 5b).
Instruct tuned versions are coming soon.
Very interesting models are coming soon too (hint: super long context).
Thanks for everyone supporting!
Pretrained on 4x more tokens than the previous releases (20b vs 5b).
Instruct tuned versions are coming soon.
Very interesting models are coming soon too (hint: super long context).
Thanks for everyone supporting!
- 3 replies
replied to their post 17 days ago
I always thought that SmolLM was one of the best SLMs around, but after the arena my thinking has changed. Maybe benchmarks aren't everything after all... BananaMind 2 Medium Chat is really impressive for its size.
replied to their post 18 days ago
I could probably have qwen 3.8 27b (or ling 3.0 tiny for speed) classify the arena data i have (around 700 battles) which would probably be enough to train a basic classifier. Letting people pick the category might be interesting for collecting training data though.
reacted to KlondikeDev's post with π₯ 18 days ago
Post
3460
A Preview of Boris-2!
Hello! Tomorrow, OpenCerebral will be releasing Boris-1.7-D60M-n30M β an experimental architecture. It will be testing a new data mixture, a new tokenizer, and testing Qwen4-like n-gram embeddings.
Following this will be Boris-1.8-D60M-n30M, which will test both the n-gram embeddings AND a new architecture.
Then, Boris-2 will begin training!
Edit: the model is OUT NOW! opencerebral/Boris-1.7-D60M-n30M
Hello! Tomorrow, OpenCerebral will be releasing Boris-1.7-D60M-n30M β an experimental architecture. It will be testing a new data mixture, a new tokenizer, and testing Qwen4-like n-gram embeddings.
Following this will be Boris-1.8-D60M-n30M, which will test both the n-gram embeddings AND a new architecture.
Then, Boris-2 will begin training!
Edit: the model is OUT NOW! opencerebral/Boris-1.7-D60M-n30M
- 71 replies
replied to their post 18 days ago
Another suggestion - and this is just a suggestion. Allow the user to select an optional category field. Default to no category. I suggest these:
- Safety
- History
- Math
- Science
- Programming
- Judgement
- Reasoning
- Culture
- Summation
- Other
Then in the leader boards, optionally allow to pick a category, defaulting to none which would render a graph including all battles.
I was almost thinking of training a small classifier model to automatically categorize the prompt, but this might be another option
reacted to appvoid's post with π₯ 19 days ago
Post
3300
Love how the small lm community is getting identity over time:
- Channel-Mixing
- XSA
- Three-tower
- Digit aware
- Loops
No one is doing the same! That's so cool.
- Channel-Mixing
- XSA
- Three-tower
- Digit aware
- Loops
No one is doing the same! That's so cool.
- 36 replies
reacted to Banaxi-Tech's post with π 20 days ago
Post
3513
We're delaying BananaMind 2.1!
When BananaMind 2.1 Lite was almost done, we benchmarked it and the results we're worse than BananaMind 2 Mini.
We're going to spend alot more time in research on tiny models and then scaling up our techniques to the actual BananaMind 2.1 models!
We're also announcing these new models:
BananaMind 2.1 Coder: A 149M instruction tuned coder model trained on 75B tokens + 10B tokens of stack-v3-train.
BananaMind 2.1 Pico: A 1M parameter model trained on 22B tokens of data.
We also may release BananaMind 2.1 Large with around 100M parameters depending on how much compute we have.
Please give us a follow!
BananaMind
@Banaxi-Tech
---
@vovaRL
@DedeProGames
When BananaMind 2.1 Lite was almost done, we benchmarked it and the results we're worse than BananaMind 2 Mini.
We're going to spend alot more time in research on tiny models and then scaling up our techniques to the actual BananaMind 2.1 models!
We're also announcing these new models:
BananaMind 2.1 Coder: A 149M instruction tuned coder model trained on 75B tokens + 10B tokens of stack-v3-train.
BananaMind 2.1 Pico: A 1M parameter model trained on 22B tokens of data.
We also may release BananaMind 2.1 Large with around 100M parameters depending on how much compute we have.
Please give us a follow!
@Banaxi-Tech
---
@vovaRL
@DedeProGames
- 7 replies
replied to their post 21 days ago
Thanks for all the suggestions! Good luck with your future models!
replied to their post 21 days ago
Haha, Iβve actually been following AI/ML developments for a year or two, and I only got seriously interested in the SLM space a couple months ago. I really only started building models myself about a month ago with MetaDiffusion though, so Iβm still learning a ton and figuring things out as I go.
replied to their post 21 days ago
Thanks! Iβm still only about a week into developing SLMs, so I still have a lot to learn.
replied to their post 21 days ago
Thatβs really generous, thank you! I think Iβd rather keep the training on my end, though. I like being able to run everything myself, and Iβd feel bad putting any pressure on you to spend your hardware and time on it. I really appreciate the offer!