Join the conversation

Join the community of Machine Learners and AI enthusiasts.

Sign Up

All HF Hub posts

Crownelius 
posted an update 1 day ago
view post
Post
4155
Howdy,
CompactAI-O is launching a tiny Model Golf, and the winner walks away with $50 in RunPod credits. Monthly. Every month. Show up, build, somebody wins.

What it is

Build the best language model you can under 100 million parameters, with at least a 1028-token context window. That's it. Any architecture, any tokenizer, any training scheme you can dream up at 3am. The only catch is it's gotta be open source (MIT, GPL, Apache, AGPL) take your pick.

It scratches the same itch as a Kaggle comp without the dataset\leaderboard nonsense. No fixed benchmark to game. No llama.cpp compatibility hoops. If you wanna train a 50M-param MoE with five experts and a tokenizer built on cookbooks, you can do that. Nothing stopping you.

The rules are listed in the discord and on the organization page if you're interested.

Why $50????

It's symbolic. It ain't gonna make anyone rich. But it's enough to cover a weekend of GPU time, enough to keep enthusiasts coming back, and not so much that it pulls in people who are just there for the money. Enthusiasts build interesting things. Interesting things move the field forward. A little incentive. I'd do it for $50 lol.

How to join

First round opens soon. Landing page is here:

CompactAI-O/Tiny-model-golf

For questions or to swap ideas, the Discord's open:

https://discord.gg/y2jTct6Cxv

Excited to see what yall come up with. ♥

— Shane
  • 8 replies
·
danielhanchen 
posted an update 1 day ago
salma-remyx 
posted an update 2 days ago
view post
Post
4950
Just trained a 2B coding model to rank candidate AI/ML research ideas against the implicit preferences in a code repository's merge history.

The training data comes from a Gaussian Process fit on the accumulated dispositions in VQASynth, where each PR against a deployed project yields a pairwise comparison between the feature branch preferred and the baseline at main.

The GP scores candidate papers to synthesize preference pairs, and DPO with LoRA bakes the ranking pipeline into the model's weights.

After 1 epoch the model reaches 87.4% reward accuracy on the held-out eval split against 92.3% on training, consistent with learning the task without overfitting.

Now, I'm scaling the pipeline to thousands of repos for a generalization test.

Dataset: remyxai/mhpd-dpo-v0
Model: remyxai/mhpd-dpo-qwen3.5-2b-vqasynth
Substack: https://remyxai.substack.com/p/the-ai-pm
kavyamanohar 
posted an update 2 days ago
view post
Post
4478
Releasing Vividh-ASR — an open benchmark and models for Hindi and Malayalam ASR.

Vividh-ASR is built from public data, stratified by complexity:
→ Clean recordings
→ Noisy and accented speech
→ Spontaneous, conversational audio

Alongside the benchmark, we release:
→ Open models for Hindi and Malayalam
→ A training recipe with two counterintuitive choices that moved the needle
→ What failed, not just what worked

The stratified evaluation methodology transfers directly to any low-resource language setup — beyond Hindi and Malayalam.

Built at @adalatai , where we build speech tech for Indian courts. This is our first open contribution back to the community. @janaab @Kush0610 @orgh0

Link: https://huggingface.co/blog/adalat-ai/vividh-benchmark
PhysiQuanty 
posted an update about 20 hours ago
Shrijanagain 
posted an update 1 day ago
view post
Post
1230
We are pleased to announce that the W-IMG Vision Dataset infrastructure is officially live.

The complete asset infrastructure is now accessible on Hugging Face for internal validation and architecture scaling targets.

Dataset Endpoint - sKT-Ai-Labs/W-IMG

#SovereignAI #ComputerVision #MachineLearning #OpenSource
kanaria007 
posted an update 1 day ago
view post
Post
98
✅ Article highlight: *Receipted World Simulation Engine* (art-60-156, v0.1)

TL;DR:
This article treats WorldSim as a *governance sandbox*.

A game already has the right shape for SI: explicit world state, discrete actions, computed effects, verification, observability, and replay. So instead of asking “is this match fair?” by vibes, WorldSim makes fairness, anti-cheat, replay fidelity, patch legitimacy, and tournament claims depend on a *receipted closed loop*.

Read:
kanaria007/agi-structural-intelligence-protocols

Why it matters:
• makes governance feel concrete and intuitive instead of abstract
• shows that “a game is SI with better UX”
• turns match fairness, replay fidelity, and anti-cheat into artifact-backed claims
• connects gameplay operations to broader SI ideas: determinism, monitoring, patch governance, publication discipline, and interop

What’s inside:
• world state as *content-addressed state* with state_ref, ticks, shards, and canonicalization
• separate *action ledgers* and *effect ledgers* so “what happened” is reconstructible
• pinned determinism + *replay receipts* for faithful replay claims
• anti-cheat framed as *adversary monitoring* with monitoring receipts
• balance patches as governed change objects with shadow apply and verification
• tournament/public statements as bounded published claims, not vibes

Key idea:
Do not say:

*“this match was fair,”*
*“this replay is faithful,”*
or *“this tournament result is official.”*

Say:

*“this result is backed by a receipted closed loop: state, actions, effects, replay, verification, publication policy, and the exact pins needed to make the claim admissible.”*
Reubencf 
posted an update 3 days ago
view post
Post
4545
I have improved my Portfolio please do check it out
Reubencf/Portfolio
  • 6 replies
·
pedrodev2026 
posted an update about 5 hours ago
prithivMLmods 
posted an update about 14 hours ago
view post
Post
148
I've made 8 Spaces in the Qwen-Image-Edit series, and out of them, 5 Spaces reached “Space of the Week”! A few Spaces are still topping the list even after many months.

Cumulatively, the series has crossed 8.2 million+ ZeroGPU runs and nearly 4 million visitors overall.

Thanks for all the community support! 🤗❤️

🔗 Spaces: https://huggingface.co/collections/prithivMLmods/image-generation-apps-collection