Hugging Face's logo

Join the conversation

Join the community of Machine Learners and AI enthusiasts.

Sign Up

All HF Hub posts
MiniMax-AI 
posted an update about 14 hours ago
view post
Post
1444
Huge news from MiniMax: we’ve secured a $2B funding round, paired with a formal long-term commitment from our CEO IO to allocate 1% of total company equity from his personal holdings to support the global open-source AI community over the next four years.

This capital backs our continuous open model releases, community tooling and transparent frontier AI research. We’re just getting started on our open-source roadmap toward accessible AGI.

If you build with open foundation models and want to push frontier AI together, come join us.
Intelligence with Everyone. 🚀

MiniMaxAI
  • 1 reply
·
sergiopaniego 
posted an update 2 days ago
view post
Post
7286
Frontier models use distillation as a step of their post-training pipelines.

In 2026 it has three jobs: compress a big model into a small one, merge RL experts into a single model, and let a model teach itself.

I wrote up which frontier models use each one and how: https://huggingface.co/blog/sergiopaniego/distillation-2026

It pairs with Class 2 of the Training an Agent series Ben and I are doing, where we teach these techniques hands-on with TRL!
  • 3 replies
·
danielhanchen 
posted an update 3 days ago
danielhanchen 
posted an update about 5 hours ago
Quazim0t0 
posted an update 1 day ago
view post
Post
3239
Disabling Gated Access for some of my models today. I will update this post with the list as I go. I had to go back recently and make updates to a lot of models and bit off more than I could chew with managing all the releases. I didn't realize many of you asked for access and I apologize for not accepting your access to the models you were wanting to look at. I don't like to release something fully unless I feel I put what I could into it for the moment. Some models will remain on gated access, but I will now be accepting those who request to view the repo.

Disabled Gated Access:
Quazim0t0/Byrne-VLM-131M - v2 Updates + Training Instructions
Quazim0t0/Byrne-Speech - 12M Tiny Speech model
Quazim0t0/Byrne-ASR-English - 12M Tiny ASR Model
Quazim0t0/Byrne-VE - Byrne-VE — Tiny Self-Distilled Vision Encoder (39M)

Accepting Gated Access Requests (7/9):
Quazim0t0/SpikeWhale-SNN-216M
Quazim0t0/Mycel-LM-79M
Quazim0t0/Chimera-64M
Quazim0t0/Positronic-144M
Quazim0t0/Wheeler-63M

Also uploaded my Neural Photonic Project:
Three trained nets in series: light interferes through the MZI2.pt optical core (verified 256/256), is measured by the PD.pt neural photodetector (verified 1024/1024), and folded into a single OUTPUT byte by the real ADC8 neural-CPU adder. Every value below is computed end-to-end by the three loaded, verified nets — no analytic formulas.
Demo: https://quazim0t0-neural-photonic-hybrid.hf.space/
Model Weights: Quazim0t0/neural-photonic

AND!

A work in progress:
Ashen Depths
https://quazim0t0-ashendepths.static.hf.space/index.html
davidmezzetti 
posted an update 3 days ago
satgeze 
posted an update 4 days ago
view post
Post
3553
First GGUF quants of Tencent's Hy3 (299B MoE), built before official llama.cpp support exists.

Hy3 dropped ~30 hours ago with only MLX and MXFP4 quants, both datacenter-sized. So I converted it myself using a community llama.cpp fork that implements the hy_v3 architecture.

What's in the repo:

- IQ1_M (62GB, fits a 128GB MacBook), IQ2_M (90GB), Q2_K (101GB), all with 1M context baked in via YaRN
- IQ quants are importance-matrix: bootstrap style. The static Q2_K ran RAM-resident to compute the imatrix, then IQ1_M and IQ2_M were requantized from the archived f16 with it
- Fixed chat template (the stock one uses .format() calls llama.cpp's Jinja rejects)
- Build instructions for the fork, including the two gotchas that cost me three build attempts

Honesty section, because that is how these repos work: this is EXPERIMENTAL. Not needle-certified yet (1M is baked but unverified, certification ladder will be published either way). MTP layer exists in the checkpoint but no llama.cpp build can run hy_v3 MTP inference yet, so it is not included. Real gate outputs are on the card, misses and all, judge for yourself.

satgeze/Hy3-1M-GGUF

Full quant ladder (Q3 through Q8) is mirroring to ModelScope for bigger hardware.
  • 9 replies
·
PhysiQuanty 
posted an update 4 days ago
view post
Post
4542
🧠 Arithmetic-SLM : A 30M model that manages to compute simple arithmetic better than a 3B model 🚀
WhirlwindAI/Arithmetic-SLM
WhirlwindAI/arithmetic-slm

🏆 Leaderboard ArithMark-2 🏆
🥇 Qwen/Qwen2.5-Math-1.5B = 82.08%
🥈 WhirlwindAI/Arithmetic-SLM = 78.60% (31.7M Params)
🥉 Qwen/Qwen2.5-3B = 78.44%

Example WhirlwindAI/Arithmetic-SLM =
0.5 * 0.5 = 0.25 ✅
105 + 45 / 8 = 110 ✅
(132 / 12) + (46 - 15) = 42 ✅
(10 + 28) * 3 = 114 ✅
1 * (16 + 28) = 44 ✅
(21 + 27) * (14 - 7) = 336 ❌ 

leaderboard = """
|              Model               |    Params    |   Score   |
|----------------------------------|--------------|-----------|
|      Qwen/Qwen2.5-Math-1.5B      |     1.54B    |   82.08%  |
|    WhirlwindAI/Arithmetic-SLM    |    31.70M    |   78.60%  | <=
|         Qwen/Qwen2.5-3B          |     3.09B    |   78.44%  |
|        Qwen/Qwen2.5-1.5B         |     1.54B    |   77.72%  |
|    Qwen/Qwen2.5-Coder-1.5B       |     1.54B    |   74.88%  |
|   HuggingFaceTB/SmolLM2-1.7B     |     1.71B    |   66.12%  |
|        Qwen/Qwen2.5-0.5B         |      494M    |   63.04%  |
| facebook/MobileLLM-R1-140M-base  |      140M    |   53.88%  |
|     SupraLabs/Supra-50M-Base     |       52M    |   27.12%  |
"""

Bench =
AxiomicLabs/ArithMark-2.0
DataSet =
WhirlwindAI/Arithmetic
By Science AND FOR SCIENCE <3
  • 3 replies
·
NatalieY 
posted an update about 2 hours ago
view post
Post
Aiden: a physical AI agent that controls phones over USB HID

Most GUI agent work assumes the agent lives inside the device or
drives it through a debugging interface. We went the other way.

Aiden is a small board that sits outside the host. It captures the
screen over HDMI-to-CSI, runs the agent loop on-device, and sends
actions back as a standard USB HID device — the host sees a keyboard
and a mouse, nothing else. No app install, no root, no ADB, no cloud.

Runtime is Go. Frame capture, full-duplex audio with VAD, the agent
loop, and HID output all run as independent goroutines. There's no
backend — nothing leaves the device, which is the only defensible
design when the input is a live feed of someone's phone screen.

Open questions we haven't solved:
· Action verification — inferring success from a re-read of the
screen breaks when loading states lie
· Prompt injection — an agent that reads screens reads whatever an
attacker puts on them
· iOS pointer control requires AssistiveTouch

Repo, including the HID gadget config and capture pipeline:
github.com/AidenAI-IO/aiden-hardware-demo

Wrote up how this differs from cloud-based computer use agents here:
https://aidenai.io/blog/mobile-ai-agent-vs-computer-use-agent-whats-the-difference/

Note: current hardware is a dev board, not a finished product.
kanaria007 
posted an update about 9 hours ago
view post
Post
18
✅ Article highlight: *Sunset and End-of-Life Governance* (art-60-200, v0.1)

TL;DR:
This article argues that “the legacy system is retired” is not enough.

Long-lived governed systems do not simply shut down. They end through live-authority closure, archive, successor handoff, retention freeze, deletion finalization, tombstone linkage, and closure receipts. 200 turns end-of-life into a first-class governance surface.

Read:
kanaria007/agi-structural-intelligence-protocols

Why it matters:
• prevents archive from becoming disappearance theater
• prevents successor handoff from laundering identity, authority, or liability
• blocks deletion-first closure while disputes, holds, or obligations remain alive
• separates ending live operation from ending governance relevance
• keeps retired systems explainable through archive bundles and tombstone linkage

What’s inside:
• end-of-life envelopes for bounded closure paths
• archive bundles for lineage, obligations, audit state, disputes, and evidence
• successor-handoff receipts for accepted and non-carried surfaces
• retention-freeze manifests for holds, deletion prerequisites, and closure criteria
• deletion-finalization receipts for what may and may not be deleted
• closure receipts for what ended, what remained, and what reentry can reopen
• tombstone-linkage records connecting retired live paths to archive and successor history

Key idea:
Do not say:

*“the old system was shut down and the new one took over.”*

Say:

*“this system entered end-of-life under this envelope, preserved this archive bundle, handed off only these admitted surfaces, froze retention before deletion, finalized only eligible deletion, emitted closure receipts, and kept tombstone linkage for future review.”*

Systems can end.

Obligations do not vanish just because the service is off.
Recently active users