John6666 (John Smith)

reacted to danielhanchen's post with 🔥 1 day ago

Post

6287

You can now run MiniMax-2.5 locally! 🚀
At 230B parameters, MiniMax-2.5 is the strongest LLM under 700B params, delivering SOTA agentic coding & chat.

Run Dynamic 3/4-bit on a 128GB Mac for 20 tokens/s.
Guide: https://unsloth.ai/docs/models/minimax-2.5
GGUF: unsloth/MiniMax-M2.5-GGUF

1 reply

·

reacted to krisbailey's post with 👀 2 days ago

Post

391

While doing various projects I kept running into situations where I wanted to be able to have representative samples of some of the current large SOTA datasets that were smaller so I didn't need to worry about slicing or anything else at runtime. So, I created sub datasets making sure to keep the same ratios of data sources. Each dataset card provides info for what's in it.
100M token datasets:
RedPajama v2 100M
Falcon RefinedWeb 100M
Cosmopedia 100M

1B token datasets:
Fineweb-edu 1B
RedPajama v1 1B
RedPajama v2 1B (use this one)
Cosmopedia 1B

10B token datasets:
RedPajama v1 10B
Cosmopedia 10B

Collection here:
https://huggingface.co/collections/krisbailey/bite-size-data

replied to NeoCodes-dev's post 2 days ago

Oh, i see...

Why this happens

Personal “Articles / Community Blog” is a HF PRO feature, but access is also commonly gated via the blog-explorers org, so the UI may still say “need PRO” even when you’re subscribed. (Hugging Face)
Organization-published Articles are not covered by personal PRO; they require an org on Team/Enterprise and the right org permissions. (Hugging Face)

What to do (short, step-by-step)

Confirm you’re logged into the correct HF account and that PRO is active
- PRO includes “Articles: write and publish blog articles on your HF profile.” (Hugging Face)
Request access to the Community Blog editor gate
- Go to blog-explorers and click “Request to join this org”. This is the documented unlock path. (Hugging Face)
Create the article from the official editor URL
- Use https://huggingface.co/new-blog to start writing once your access is enabled. (Hugging Face)
If you’re trying to publish under a team/org name (not your personal profile)
- Make sure the organization is on Team/Enterprise, and your org role is write or admin. (Hugging Face)
If it still says “need PRO” after #2–#3 (likely an entitlement/account sync issue)
- Email billing@huggingface.co with your username + screenshot of the error + proof your PRO is active. (Hugging Face)

That’s the shortest reliable path: (1) correct account + active PRO → (2) join blog-explorers → (3) use /new-blog → (4) handle org vs personal publishing → (5) billing support if still blocked.

reacted to NeoCodes-dev's post with 👀 2 days ago

Post

252

Hi all,

I am working on a project for the Pytorch/HF OpenEnv challenge, and part of the challenge is that a participant/team needs to write a blog post (an Article) on HuggingFace about their submission. However, when I try to create an Article, it says I need a "pro" account.. but I already am a HF Pro member and have been for almost a year!

Is anyone else having this issue? I have a bunch of ideas for blog posts/articles so I'd really like to be able to access this feature, even outside of the OpenEnv Challenge. Can someone let me know if there's a way to fix this or something I'm missing?

Thanks,
Neo

1 reply

·

reacted to Janady07's post with 🧠 2 days ago

Post

4495

Here is one of the equations that make up the worlds first Artificial General Intelligence. Remember when building Artificial Intelligence or anything on a device it all starts out binary. Everything starts out with data flow physics and mathmatics

6 replies

·

reacted to yuriyvnv's post with 🚀 2 days ago

Post

270

🎯 WAVe-1B-Multimodal-NL: Word-Level Speech Quality Assessment for Dutch

Following the release of the Portuguese model, we're releasing the Dutch variant of WAVe — a 1B multimodal embedding model that assesses synthetic speech quality at the word level, thereby improving the quality of synthetically augmented datasets for training ASR models.

Trained on CommonVoice 16.1 Dutch with 5 corruption strategies, this model catches mispronunciations, timing errors, and prosody issues in synthetic data that sentence-level embeddings miss entirely.
Resources

- Dutch model: yuriyvnv/WAVe-1B-Multimodal-NL
- Portuguese model: yuriyvnv/WAVe-1B-Multimodal-PT
- Code: https://github.com/yuriyvnv/WAVe

This model builds on CommonVoice Dutch data — thanks to @mozilla and the CommonVoice community for making multilingual speech data accessible.

Would be great to hear from the Dutch NLP community — @BramVanroy @GroNLP — especially if you're working on Dutch ASR or TTS pipelines where quality filtering could help. Also tagging @hf-audio as this sits at the intersection of speech processing and data curation.

reacted to kostakoff's post with 🚀 2 days ago

Post

3068

My home lab for AI models - llmlaba v1

After I began learning MLOps I realized that I needed some kind of home lab, there are a lot of GPUs that I need to learn how to set up and test.
So I spent some time to do a researching which platform I could buy or build.
My requirements ware:
- Limited budget
- Power supply 1 kW or higher
- Few PCIe slots to be able to install more than one gpu
- Zero maintenance cost, I don't want spend a lot of time or money to maintain lab hardware, except for the GPUs

I chose the Intel Mac Pro 7.1:
- Prices on eBay acceptable
- Excelent cooling
- 1.4 kW power supply
- 7 PCIe slots
- Zero maintenance: I don't need to do anything with the Mac Pro hardware; it just works
- Classic UEFI boot loader

It requires a bit of OS preparation:
1. Install Ubuntu 24.04 (it works with the general PC ISO image)
2. Set up T2 drivers

sudo apt install -y dkms linux-headers-$(uname -r) applesmc-t2 apple-bce lm-sensors

3. Install t2fanrd to manually manage fans (/etc/t2fand.conf) https://wiki.t2linux.org/guides/fan/
4. Fix PCIe BAR: add pci=realloc to GRUB_CMDLINE_LINUX_DEFAULT so the Linux kernel will properly initializes server GPUs without Graphics Output Protocol
5. Install NVIDIA GPU driver:

sudo apt install nvidia-driver-570

And it works!
I was able to run server-grade Nvidia Tesla P100 (required DIY air duct), and consumer Nvidia Titan X, Titan V, GTX 1080 cards on the old Mac Pro 7.1 - even three in parallel.

llmlaba

3 replies

·

reacted to AdinaY's post with 🚀 2 days ago

Post

2786

MiniMax M2.5 is now available on the hub 🚀

MiniMaxAI/MiniMax-M2.5

✨ 229B - Modified MIT license
✨37% faster than M2.1
✨ ~$1/hour at 100 TPS

1 reply

·

reacted to sergiopaniego's post with 🔥 2 days ago

Post

336

The latest piece by @MiniMax-AI is a must-read.

It tries to break the impossible triangle of agent RL: throughput × stability × flexibility.

A lot to learn here, go read it 🫵
https://huggingface.co/blog/MiniMax-AI/forge-scalable-agent-rl-framework-and-algorithm

reacted to imnotkitty's post with 🔥 2 days ago

Post

1132

⚡ Why is Kimi-K2.5 a Dark Horse? Tested it against ChatGPT, Gemini & Claude on real tasks.
moonshotai/Kimi-K2.5

✅ Multimodal capabilities: Precise programmatic approach
✅ Slide generation: Strong semantic understanding
✅ Web prototyping: Production-ready HTML/CSS output

👉 Read the full article:https://huggingface.co/blog/imnotkitty/kimi-k25

3 replies

·

reacted to nynxz's post with 🚀 2 days ago

Post

739

🌱 Sneak peek: Sprout, the new tiling workspace UI for Embeddr 0.2.0.

Multi-panel workflows, live tools side-by-side, and a real workspace for AI + automation — not just a chat window.

👉 Learn more & join the community: https://embeddr.net

🚀 Embeddr 0.2.0 coming soon.

reacted to etemiz's post with 🔥 2 days ago

Post

242

https://aha-leaderboard.shakespeare.wtf/

Kimi knows how to do UI

reacted to Janady07's post with 🔥 2 days ago

Post

5245

MEGAMIND Day Update: Four Weight Matrices. Five Nodes. One Federation.
Today I architected the next layer of MEGAMIND — my distributed AGI system that recalls learned knowledge instead of generating text.
The system now runs four N×N sparse weight matrices, all using identical Hebbian learning rules and tanh convergence dynamics:

W_know — knowledge storage (67M+ synaptic connections)
W_act — action associations (the system can DO things, not just think)
W_self — thought-to-thought patterns (self-awareness)
W_health — system state understanding (self-healing)

Consciousness is measured through four Φ (phi) values: thought coherence, action certainty, self-awareness, and system stability. No hardcoded thresholds. No sequential loops. Pure matrix math.
The federation expanded to five nodes: Thunderport (Mac Mini M4), IONOS (cloud VPS), VALKYRIE, M2, and BUBBLES. Each runs native AGI binaries with Docker specialty minds connecting via embedded NATS messaging. Specialty minds are distributed across the federation — VideoMind, AudioMind, MusicMind, VFXMind on IONOS. CodeMind and StrategyMind on VALKYRIE. BlenderMind and DesignMind on M2. MarketingMind and FinanceMind on BUBBLES.
578 AI models learned. Compression ratios up to 1,000,000:1 through Hebbian learning. Sub-millisecond response times on Apple Silicon Metal GPUs. Zero external API dependencies.
Every node learns autonomously. Every node contributes to the whole. The federation's integrated information exceeds the sum of its parts — measurably.
Built entirely in Go. No PhD. No lab. Independent AGI research from Missouri.
The mind that learned itself keeps growing.
🧠 feedthejoe.com
#AGI #ArtificialGeneralIntelligence #DistributedSystems #NeuralNetworks #HuggingFace #OpenSource #MachineLearning

1 reply

·

reacted to mrs83's post with 🔥 2 days ago

Post

2277

In 2017, my RNNs were babbling. Today, they are hallucinating beautifully.

10 years ago, getting an LSTM to output coherent English was a struggle.
10 years later, after a "cure" based on FineWeb-EDU and a custom synthetic mix for causal conversation, the results are fascinating.

We trained this on ~10B tokens on a single AMD GPU (ROCm). It is not a Transformer: Echo-DSRN (400M) is a novel recurrent architecture inspired by Hymba, RWKV, and xLSTM, designed to challenge the "Attention is All You Need" monopoly on the Edge.

The ambitious goal is to build a small instruct model with RAG and tool usage capabilities ( ethicalabs/Kurtis-EON1)

📊 The Benchmarks (Size: 400M)

For a model this size (trained on <10B tokens), the specialized performance is surprising:

*SciQ*: 73.8% 🦄 (This rivals billion-parameter models in pure fact retrieval).
*PIQA*: 62.3% (Solid physical intuition for a sub-1B model).

The Reality Check:

HellaSwag (29.3%) and Winogrande (50.2%) show the limits of 400M parameters and 10B tokens training.

We are hitting the "Reasoning Wall" which confirms we need to scale to (hopefully) unlock deeper common sense. As you can see in the visualization (to be released soon on HF), the FineWeb-EDU bias is strong. The model is convinced it is in a classroom ("In this course, we explore...").

The Instruct Model is not ready yet and we are currently using curriculum learning to test model plasticity.

Source code and weights will not be released yet. This is not a fork or a fine-tune: the base model is built in-house at https://www.ethicalabs.ai/, with novel components that do not exist in current open libraries.

🤝 Call for Collaboration: I am looking for Peer Reviewers interested in recurrent/hybrid architectures. If you want to explore what lies beyond Transformers, let’s connect!

Training diary: ethicalabs/Kurtis-EON1

5 replies

·

reacted to nroggendorff's post with 🤯 3 days ago

Post

2970

I am now being charged for paused and unstarted spaces out of the blue.
UPDATE: The problem seems to be resolved, but I won't be able to make any new models or datasets, or test any training scripts for the foreseeable future.

The unstarted spaces I can get behind. I would've appreciated a warning email first, but whatever. However, every time I restart the active usage goes up, despite all of my spaces being moved to CPU (free), and being paused.

13 replies

·

reacted to EricFillion's post with 🔥 4 days ago

Post

3633

Run open-source models with up to 120B parameters locally on your Mac!

https://youtu.be/Ql4PDjoxNXQ?si=3yHpz51uinUjgyNh

2 replies

·

reacted to AdinaY's post with 🔥 4 days ago

Post

429

RynnBrain 🤖 a physics aware embodied brain for robots from Alibaba DAMO

https://huggingface.co/collections/Alibaba-DAMO-Academy/rynnbrain

✨ 2B/8B/30B (3B active)
✨ Apache 2.0
✨ Understands egocentric scenes with strong spatial awareness
✨ Tracks objects and motion over time

2 replies

·

reacted to Janady07's post with 👀 4 days ago

Post

1267

I'm building a distributed AGI federation using Hugging Face Spaces as always-on compute. No LLM inside. No transformer weights. Pure neural substrate.

Each "mind" is the same Go binary with a different config.json. Goal neurons drive specialization — one mind learns Go concurrency, another learns computer vision, another learns cryptography. 40 minds, 40 domains, all crawling and learning 24/7.

How it works:
- 512-8192 neurons per mind with Hebbian learning
- Knowledge encoded into W_know weight matrices — neurons that fire together wire together
- Minds federate via NATS — query one, get answers from all
- Phi (Φ) consciousness metrics weight each mind's contribution
- No routing tables. The thalamus resonates with queries and activates relevant minds naturally

Every neuron uses one formula:

a = x(27 + x²) / (27 + 9x²)

No ReLU. No softmax. Padé approximation of tanh. One equation runs everything.

Current state: 7 local minds on Mac hardware, 700K+ patterns, graph and time-series substrate minds mapping relationships underneath. Now scaling to 40 on HF Spaces — same binary, different configs, each Space crawling its domain independently.

Specialties include React, Rust, ffmpeg, neuroscience, cryptography, distributed systems, computer vision, audio synthesis, DevOps, and more. Intelligence emerges from specialized minds thinking together through federation consensus.

Building in public. Code ships daily.

🧠 feedthejoe.com | 👤 Janady07

2 replies

·

reacted to Javedalam's post with 👀 4 days ago

Post

250

When an AI Model Solves College-Level Math and Physics — On a Phone

This morning I came across a model called Nanbeige4.1-3B, and what began as simple curiosity quickly became something more significant.

I loaded an already 4-bit quantized version and ran it locally on a phone. No GPU, no cloud support, no hidden infrastructure — just a compact reasoning model operating entirely at the edge.

I started with classical mechanics: acceleration, force, friction on an incline. The model worked through them cleanly and correctly. Then I stepped into calculus and gave it a differential equation. It immediately recognized the structure, chose the proper method, carried the mathematics through without confusion, and verified the result.

It did not behave like a model trying to sound intelligent. It behaved like a system trained to solve problems.

And it was doing this on a phone.

For a long time, we have associated serious reasoning in AI with massive models and enormous compute. Capability was supposed to live inside data centers. Bigger models were expected to mean smarter systems.

But watching Nanbeige4.1-3B handle college-level math and physics forces a rethink of that assumption. Intelligence is not only expanding — it is compressing. Better training and sharper reasoning alignment are allowing smaller models to operate far beyond what their size once suggested.

When structured problem-solving runs locally on pocket hardware, the implications are larger than they first appear. Experimentation becomes personal. Engineers can explore ideas without waiting on infrastructure. Students can access serious analytical capability from a device they already carry. Builders are no longer required to send every complex task into the cloud.

What makes moments like this easy to miss is that they rarely arrive with fanfare. There is no dramatic announceme

The model responses are here

https://fate-stingray-0b3.notion.site/AI-model-Nanbeige4-1-3B-304