Building whryte.com, Creator of localGPT | AI Educator
- Tweets 1,198
- Following 1,404
- Followers 2,357
- Likes 3,343
Pinned Tweet
Yesterday I released the 'preview' of LocalGPT v2, and its already trending on Github
Its an opinionated implementation of private RAG powered by local models via @ollama and @huggingface. Give it a ⭐️ on @github (🙏🙏)
Watch the video in next post to learn how it was built...
3
2
27
2,875
I have been testing this in early access and it's really impressive, specially for real-time streaming audio. The best part, it's entirely local!
When several people talk at once, a transcript can get messy fast.
Our new Nemotron 3 Diarization model tracks who spoke when, even when voices overlap. It handles up to eight speakers, has 100M parameters, and is now available on @huggingface 🤗
6
216
What a day so far. Anyone else planning on dropping another model :)
Please welcome GPT-6 Sol and GPT-6 Luna to the GPT-6 universe.
GPT-6 Sol and Luna build on the advances behind GPT-6 Astra, bringing much of its strengths into faster and more affordable models to support work at scale.
We’ve also made caching and inference more efficient, and we’re passing the savings directly to you: 50% lower API prices for Sol and Luna compared with GPT‑5.6 promotional pricing.
1
1
181
This is weird benchmark behavior from Claude models (not just Opus 5.5). Not sure what is happening.
2
1
261
Union Alpha: This model is surprisingly great for UI design. I wasn't expecting this quality with a relatively simple prompt.
Tends to follow prompts really closely in my quick tests.
Will be sharing more results here. So far I am impressed.
Might make a video on it.
🥷 New stealth model: Union Alpha (@unionalphaai)
A multimodal model for research, coding, and agentic workflows.
- Free to use
- 256K context
- Tool calling
- Frontier-level general-purpose performance
Try it now and share your feedback:
openrouter.ai/stealth/union-…
1
283
Prompt retweeted
This isn't the most notable aspect of today's news, but on the user data issue, there are different kinds of *training on user data* with very different privacy/IP implications. Sadly, AI cos don't like to disclose what they're doing.
- pretrain on user data, with users' tokens as prediction targets: high regurgitation risk, improper
- use user prompts to distill large models into small ones: low regurg. risk, some companies probably do this
- use user traces to construct RL tasks: low regurg. risk, because RL has low memorization abilities, but can extract customer IP, depending on how it's done. Ranges from benign "use explicit user feedback in reward model training" to invasive "upload user's coding environment and commit history to turn into rl envs"
"De-identification" is weak -- you can identify someone with a small number of bits, and long traces have more than enough. And it doesn't affect IP leakage concerns.
Two things to distinguish:
Did any human or agent look at user data as part of the Navier Stokes effort? No.
Do we use user feedback and de-identified data to improve ChatGPT and Codex in a holistic way? Yes. And so does every LLM company.
44
154
1,360
281,532
Acceleration!
We’re sharing a solution to the Navier-Stokes Millennium Prize Problem, one of the deepest problems at the frontier of mathematics.
The proof was produced by a group of agents, using an OpenAI next-generation model significantly more capable than GPT-6 Astra.
The problem concerns whether the description of smooth three-dimensional fluid motion modeled by the Navier-Stokes equations can break down. It has remained unresolved for roughly 90 years.
136
Its official!
Exciting day for NVIDIA and @huggingface.
Open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty. They allow every developer, startup, university, industry and country to build with, customize and benefit from AI.
Thank you @ClementDelangue for coming to me.
NVIDIA is going to be a great home for Hugging Face, its community and the future of open models. 🤗
blogs.nvidia.com/blog/nvidia…
1
6
353
This is pretty amazing, I was one of the early testers and its really useful for educational content creators along with others.
It can actually understand what is happening in the video and can really provide a detailed feedback on what to improve visually or delivery/style.
Agentic video understanding is available now across our latest models via the Gemini API in @GoogleAIStudio and the Gemini Enterprise Agent Platform.
Coming soon to the @GeminiApp and to @YouTube's “Ask YouTube” feature on the video watch page.
Learn more ↓ goo.gle/4x5Knd1
5
401
Your Agent needs it's own computer!
Really excited for this launch!
Started at an @AnthropicAI Hackathon in February.
Today, Matrix OS (@joinmatrixos) is live, giving a computer for your AI agents so they don't die when you close your laptop.
But for me, the story started much earlier.
When I was six, I could barely read, but I would spend hours clicking through Windows XP just trying to understand what a computer could do. I think I’ve been obsessed with that question ever since.
Over the past 3.5 years, I’ve built intensively with AI. I kept seeing the same pattern: the models were getting incredibly capable, but the computer around them was still designed for a human sitting in front of a screen, clicking around and typing.
I joined forces with @NimaNaderi2004, we kept building, and today we’re opening signups, shipping Matrix and our desktop app, and announcing our first angel round, backed by @alexwikstrom, @bakkermichiel, @salar, @Founder_Ash, @mattiasmiksche, @fbjork, @sidk_, and @engineerrprompt among others.
Since we are backed by the best, we brought @xyhstruggler , @Elibethmarch and @valeriewistrom to help build and achieve one mission.
Our mission is to give everyone a Personal Cloud Computer 💚
PS: We’re also hiring engineers in Stockholm and around the world. If rethinking what a computer can be; sounds like your kind of problem, come build with us!
1
2
678
OX Alpha (GLM 5.3 Flash) is out and it looks pretty good!
Happy GLM 5.3 Flash day to all who celebrate!
320B @ 18B, and fighting off the giants. This model is truly awesome.
@Zai_org are one the best 🥺
huggingface.co/zai-org/GLM-5…
1
3
203
this is awesome! we need more support for local models.
Junie can now run entirely on your machine, and it’s completely free to use.
One command: /local. We picked and optimized the whole stack around the model: agent harness, model parameters, and inference engine. No tokens, no quota, no code ever leaving your machine.
7
347
Some experiments with OX Alpha on 9 different harnesses.
- Scores ranged from 21/40 to 37/40.
- Token use per task varied 6x.
- Setup: OX Alpha inside pi, hermes, kimi, opencode, cline, claude code, codex, prime and omp.
- 40 agentic coding tasks: 20 standard + 20 expert-level.
- Deterministic pass/fail checkers, no LLM judging.
9
2
30
3,013
Hard tasks rewarded persistence.
One hermes solve took 98 API calls and 43 minutes of steady tool use.
claude ended its expert runs after ~4.5K generated tokens. The same decisive style that made it 20/20 on standard work left hard problems unfinished.
1
2
171
The important caveat: this is NOT a harness ranking.
It measures how well each harness fits THIS model. The claude lane scored 59/60 with a different model on this exact suite.
Swap the model and the ordering will probably change. The pairing is what you should measure.
Full breakdown with all the charts (494 runs, 123M prompt tokens, total API cost $0):
engineerprompt.ai/writing/ox…
3
161