K-ai-Innovations (Jeremy Hill)

reacted to ZennyKenny's post with 🚀 about 12 hours ago

Post

1732

😎 My new personal website is live! Check out https://kennethhamilton.me to chat with an LLM about my professional skills and personal projects.

🙈 Think of it like a really, really vain version of ChatGPT.

6 replies

·

reacted to mmhamdy's post with 🚀🔥 5 days ago

Post

2958

The new DeepSeek Engram paper is super fun! It also integrates mHC, and I suspect they're probably releasing all these papers to make the V4 report of reasonable length😄

Here's a nice short summary from Gemini

liked a model 7 days ago

Valdemardi/Dolphin3.0-R1-Mistral-24B-AWQ

Text Generation • 24B • Updated Feb 7, 2025 • 130 • 6

reacted to unmodeled-tyler's post with 🚀 16 days ago

Post

1388

Happy New Year, Hugging Face!

It's been a crazy year for me! This year I launched VANTA Research as a solo operator and managed to push out 14 original open source finetunes and 5 datasets in the span of about 4 months, completely on my own.

The reception has been much higher than I ever anticipated and sincerely appreciate everyone that's checked out my work thus far.

The good news is, I'm just getting started! In 2026 you can expect even more original models from VANTA Research, more open source datasets, and maybe some other cool things as well? 👀

2026 is gonna be big for AI in general, and I can't wait to experience it with all of you!

1 reply

·

liked a dataset 20 days ago

kanaria007/agi-structural-intelligence-protocols

Updated 6 days ago • 471 • 8

upvoted an article 20 days ago

Article

Deep-Space SI-Core: Autonomy Across Light-Hours - How an onboard SI-Core evolves safely while Earth is hours away

22 days ago

•

2

reacted to kanaria007's post with 🚀 20 days ago

Post

1868

✅ New Article: *Deep-Space SI-Core — Autonomy Across Light-Hours*

Title:
🚀 Deep-Space SI-Core: Autonomy Across Light-Hours - How an onboard SI-Core evolves safely while Earth is hours away
🔗 https://huggingface.co/blog/kanaria007/deep-space-si-core

---

Summary:
Most autonomy stories quietly assume “someone can intervene in minutes.” Deep space breaks that assumption.
With 2–6 hours round-trip latency and intermittent links, an onboard SI-Core must act as a *local sovereign*—while remaining *globally accountable* to Earth.

This note sketches how mission continuity survives when nobody is listening: DTN-style semantic bundles, local vs. global rollback, bounded self-improvement, and auditability that still works after contact windows return.

> Autonomy isn’t a divorce from governance—
> it’s a measured loan of authority, under a constitution, with evidence.

---

Why It Matters:
• Makes “autonomous” mean *operational*, not rhetorical, under light-hour delays
• Clarifies how rollback works when you can’t undo physics—only *policy trajectories*
• Shows how an onboard core can *self-improve without drifting out of spec*
• Treats *silence itself as an observation* (missing logs are governance signals)

---

What’s Inside:
• Two-core model: *Earth-Core (constitutional/strategic)* vs *Ship-Core (tactical/operational)*
• *SCP over DTN* as semantic bundles (priorities, idempotency, meaning checkpoints)
• Local rollback vs. epoch-level governance (“retroactive” steering without pretending to reverse time)
• Bounded onboard learning + LearningTrace for later audit and resync
• Stress scenario walkthrough: micrometeoroid storm, compound failures, and graceful degradation
• Metrics framing for deep space: governability, audit completeness, ethics uptime, rollback integrity

---

📖 Structured Intelligence Engineering Series

reacted to csabakecskemeti's post with 🔥 23 days ago

Post

3192

Just sharing a result of a homelab infrastructure experiment:

I've managed to setup a distributed inference infra at home using a DGX Spark (128GB unified gddr6) and a linux workstation with an RTX 6000 Pro (96GB gddr7) connected via 100Gbps RoCEv2. The model I've used (https://lnkd.in/gx6J7YuB) is about 140GB so could not fit either of the GPU. Full setup and tutorial soon on devquasar.com

Screen recording:
https://lnkd.in/gKM9H5GJ

3 replies

·

reacted to samerzaher80's post with 🚀 about 2 months ago

Post

1854

Need Help Getting arXiv Endorsement for My AI Research Paper

Hi everyone,
I hope you're doing well. I’m trying to publish my new AI research paper on arXiv under the cs.AI category, but I currently need an endorser who is already authorized for cs.AI submissions.

If anyone here is registered as a cs.AI endorser and is willing to help, I would truly appreciate it.

Here is the official arXiv endorsement request link:

🔗 https://arxiv.org/auth/endorse?x=EZEMO7
(Backup: http://arxiv.org/auth/endorse.php — Code: EZEMO7)

My research:
It’s part of the AetherMind project — a self-reflective NLI reasoning system inspired by human cognitive consistency and used also in Alzheimer’s research. If needed, I can share the abstract or full PDF.

Thank you so much to anyone who can support.

— Sameer S.Najm

reacted to nouamanetazi's post with 👍🤗🚀 3 months ago

Post

4300

After training 𝐒𝐦𝐨𝐥𝐋𝐌𝟑 on 𝟑𝟖𝟒 𝐇𝟏𝟎𝟎𝐬 for nearly a month, I've come to realize something most people overlook: 𝐢𝐧𝐟𝐫𝐚𝐬𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞 𝐢𝐬 𝐭𝐡𝐞 𝐦𝐚𝐤𝐞-𝐨𝐫-𝐛𝐫𝐞𝐚𝐤 𝐟𝐚𝐜𝐭𝐨𝐫 𝐢𝐧 𝐋𝐋𝐌 𝐭𝐫𝐚𝐢𝐧𝐢𝐧𝐠. 🔥

Everyone talks about model architecture and data quality. And yes, those matter immensely. But here's what nobody tells you: when your training run fails at 2 AM because of mysterious 𝐍𝐂𝐂𝐋 𝐞𝐫𝐫𝐨𝐫𝐬, or when your expensive GPU cluster is running at 𝟔𝟎% 𝐞𝐟𝐟𝐢𝐜𝐢𝐞𝐧𝐜𝐲, the problem isn't your model. It's most probably a 𝐦𝐢𝐬𝐮𝐬𝐞 𝐨𝐟 𝐭𝐡𝐞 𝐡𝐚𝐫𝐝𝐰𝐚𝐫𝐞. 🛠️

Questions that seemed simple but had no clear answers: Why is 𝐌𝐨𝐄 𝐭𝐫𝐚𝐢𝐧𝐢𝐧𝐠 𝐬𝐥𝐨𝐰𝐞𝐫 𝐭𝐡𝐚𝐧 𝐝𝐞𝐧𝐬𝐞 𝐦𝐨𝐝𝐞𝐥𝐬? Which 𝐍𝐂𝐂𝐋 𝐟𝐥𝐚𝐠𝐬 should we actually set? How often should we checkpoint without killing throughput?

That's why we built 𝐓𝐡𝐞 𝐒𝐦𝐨𝐥 𝐓𝐫𝐚𝐢𝐧𝐢𝐧𝐠 𝐏𝐥𝐚𝐲𝐛𝐨𝐨𝐤 📖: a complete guide covering everything from model architecture and data curation to the SmolLM3 training marathon, post-training techniques, and crucially, the 𝐢𝐧𝐟𝐫𝐚𝐬𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞 𝐥𝐚𝐲𝐞𝐫 that most teams get wrong.

We validated real vs theoretical bandwidth across the entire stack: 𝐇𝐁𝐌𝟑 𝐡𝐢𝐭𝐭𝐢𝐧𝐠 𝟑 𝐓𝐁/𝐬, 𝐍𝐕𝐋𝐢𝐧𝐤 𝟒.𝟎 𝐫𝐞𝐚𝐜𝐡𝐢𝐧𝐠 𝟕𝟖𝟔 𝐆𝐁/𝐬, 𝐏𝐂𝐈𝐞 𝐆𝐞𝐧𝟒 𝐚𝐭 𝟏𝟒.𝟐 𝐆𝐁/𝐬. Then we ran collective operations across 𝟏𝟐𝟖 𝐆𝐏𝐔𝐬 (16 nodes, 8xH100s each) and measured how performance degrades at scale: all-reduce drops from 𝟒𝟖𝟎 𝐆𝐁/𝐬 on a single node to 𝟑𝟐𝟎-𝟑𝟓𝟎 𝐆𝐁/𝐬 across 16 nodes.

If you've ever wondered why your training runs are slower than they should be, or you're planning to scale up and want to avoid expensive mistakes, this guide might save you weeks of debugging.

𝐓𝐡𝐞 𝐒𝐦𝐨𝐥 𝐓𝐫𝐚𝐢𝐧𝐢𝐧𝐠 𝐏𝐥𝐚𝐲𝐛𝐨𝐨𝐤: https://lnkd.in/e5MKXUHS

Shared with ❤️ by the HuggingFace team

liked a model 3 months ago

racineai/QwenAmann-4B-dse

Visual Document Retrieval • 4B • Updated Oct 19, 2025 • 6 • 15

upvoted an article 3 months ago

Article

How I Built Lightning-Fast Vector Search for Legal Documents

Oct 20, 2025

•

14

reacted to adlumal's post with 👍 3 months ago

Post

2433

I benchmarked embedding APIs for speed, compared local vs hosted models, and tuned USearch for sub-millisecond retrieval on 143k chunks using only CPU. The post walks through the results, trade-offs, and what I learned about embedding API terms of service.
The main motivation for using USearch is that CPU compute is cheap and easy to scale.

Blog post: https://huggingface.co/blog/adlumal/lightning-fast-vector-search-for-legal-documents

2 replies

·

reacted to unmodeled-tyler's post with 😎 3 months ago

Post

1970

I just set up the new Ollama integration in VS Code, so I wanted to test it. I hooked up glm-4.6, and asked it to build a full stack Ollama chat interface.

In only 3 prompts, glm-4.6 built the app, and debugged it successfully. One prompt for the build, two for debugging -> fully functional app.

I was genuinely impressed! It's really cool to see how powerful open source tools have become. The future is exciting and I'm here for it!

reacted to mike-ravkine's post with 🔥❤️ 3 months ago

Post

3751

Let's talk about one of the hidden gems in the ReasonScape evaluation results, lucky #13: https://huggingface.co/aquif-ai/aquif-3.5-8B-Think

Built on top of the solid Qwen3-8B foundation, aquif-3.5-8B-Think successfully preserves the high performance of the original model while consuming 30-50% less reasoning tokens.

The most notable regression vs the base model here is in arithmetic - if your workload is math heavy this model demonstrates an unfortunate collapse with performance under growing complexity.

The interesting combination of awesome overall performance on SVG simple shapes identification coupled with a total inability to recognize more complex shapes like 'House' or 'Arrow' is a behavior directly inherited from the base model (but with a ~20% improvement in token utilization).

If you like your reasoning models token-efficient, Aquif-3.5-8B-Think is well worth a spin.

Higher resolution, more detailed, interactive plots are available at the m12X explorer: https://reasonscape.com/m12x/explorer/

1 reply

·

reacted to unmodeled-tyler's post with 🔥 3 months ago

Post

3688

vanta-research/apollo-astralis-8b

I ran the same prompt sequence on my model Apollo Astralis 8B and Hermes4 14B from Nous Research.. The raw chat logs were then given to 3 different architectures (DeepSeek 3.1, LLaMA 405B, GPT-OSS 120B). All 3 models were given the same, simple instructions to analyze the logs and determine which model performed better.

All 3 independently chose Astralis 8B for stronger reasoning, alignment, transparency, and collaborative language.

Astralis 8B is designed to keep you motivated by applying warm collaborative language mixed with rigorous logical reasoning and problem solving capabilities.

Give Astralis a try!

2 replies

·

Jeremy Hill

AI & ML interests

Recent Activity

Organizations

Valdemardi/Dolphin3.0-R1-Mistral-24B-AWQ

kanaria007/agi-structural-intelligence-protocols

Deep-Space SI-Core: Autonomy Across Light-Hours - How an onboard SI-Core evolves safely while Earth is hours away

racineai/QwenAmann-4B-dse

How I Built Lightning-Fast Vector Search for Legal Documents

Jeremy Hill

AI & ML interests

Recent Activity

Organizations

K-ai-Innovations's activity

Deep-Space SI-Core: Autonomy Across Light-Hours - *How an onboard SI-Core evolves safely while Earth is hours away*

How I Built Lightning-Fast Vector Search for Legal Documents

Deep-Space SI-Core: Autonomy Across Light-Hours - How an onboard SI-Core evolves safely while Earth is hours away