🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Smart Turn v2: open source, native audio turn detection in 14 languages. New checkpoint of the open source, open data, open training code, semantic VAD model on Hugging Face, fal, and Pipecat AI. - 3x faster inference (12ms on an L40) - 14 languages (13 more than v1, which

you've been asking for on how to catch up with new models, ok revealing a life hack I use Hugging Face MCP a lot, the prompt: "what are the trending models datasets and spaces for past week, created past week on Hugging Face? bring 10 for each, their model size, details"

Aaj humara janamdin hai

Quarter life : Itna bhi mature nahi hona tha ki apne bday pe dusro ko wishes de rahe ho

Something wong is obviously brewing for tech landscape of India or I must say how our R and D is headed First Sama meeting A R rahman and then Aarvind , Tech people of US meeting Indian bollywood means only one thing and that is finding that right disturbution channel to a

It’s either so over or we’ve never been so back

Can we all take a moment to disagree and Say 🦀 is better than 🐍

x.com/i/spaces/1YqJD…

Local voice AI with a 235 billion parameter LLM. ✅ - smart-turn v2 - MLX Whisper (large-v3-turbo-q4) - Qwen3-235B-A22B-Instruct-2507-3bit-DWQ - Kokoro All models running local on an M4 mac. Max RAM usage ~110GB. Voice-to-voice latency is ~950ms. There are a couple of

I don't know but I feel really happy reading this today have been with him for almost a year now and all of a sudden I am reading things I have absolutely no shiiting idea about like MTNL and DNS what as vibhor truly said you shouldn't stop tinkering with ideas you are

I have been an adamant bielever in education cannot be fostered by locking young talents in modern brick jails it finally boils down to talent leaving the country or making sure the next generation does Her incident gives me chills, reminds me of my time in college where there

Graph-R1 New RAG framework just dropped! Combines agents, GraphRAG, and RL. Here are my notes:

If and only if TCS was like Tata Motors we would all have been at a better offset position

Here's a F.02 in my home, using Helix to do my laundry

Gautam Kumar Linkedin - Use Boolean search : linkedin.com/help/linkedin/… WeWorkRemotely RemoteOK Wellfound remotive.com RemoteIndian Slack community Proxify AB - gets you global clients - Talent Network. Joined this recently. Turing - Now only LLM training ad less dev jobs Upwork -

Researchers designed a 27M model inspired by the brain. It outperforms o3-mini, R1, and others. This makes every single AI lab look silly! Here are their 5 core techniques: - Brain-inspired hierarchical processing, temporal separation, and recurrent connectivity - Hierarchical

Introducing Kitten TTS, a SOTA tiny text-to-speech model - Just 15M parameters - Runs without a GPU - Model size less than 25 MB - Multiple high-quality voices - Ultra-fast - even runs on low-end edge devices Github and HF links below

congratulations Sam Altman on the amazing release of gpt-oss 🥳

Booked AIR INDIA for IIT Rohini via IIT seelampur

Yeah I slept again early last night and GPT went full OSS Google launches Genie Claude went Opus 4.1 Someone IndieHacker released a near realtime TTS which even I am working on Sed lyf I need to bring down my sleep more