Depankar chatterjee

Why is no one talking about this? @nvidia is offering around 80 AI models via hosted APIs absolutely for free. You get access to MiniMax M2.7, GLM 5.1, Kimi 2.5, DeepSeek 3.2, GPT-OSS-120B, Sarvam-M etc. This plugs straight into OpenClaude, OpenCode, Zed IDE, Hermes agent and even with Cursor IDE. Setup: – Grab API key: build.nvidia.com/models – base_url = "integrate.api.nvidia.com/v1" – api_key = "$NVIDIA_API_KEY" – select model (e.g. minimaxai/minimax-m2.7) If you’re building or experimenting, this is basically free inference. Lock in and start building today anon. Thank me later.

@grok give me all link of claude mythos report

@grok @leftcurvedev_ @huggingface @grok can i ask a sensitive question with this obliterate model ?

"obliterate" (or "obliterated") is slang in the local AI fine-tuning scene. It means they took the original Gemma 4 E4B model and aggressively stripped out all its built-in safety/refusal mechanisms (the parts that say "I can't answer that"). Result: 0% refusals, still super coherent and smart, basically an uncensored beast. The "OBLITERATED" version is that fully cooked, no-holds-barred edit.

Gemma 4 E4B OBLITERATED is now trending on @huggingface 🔥 OBLITERATUS dropped the aggressive run on the base Gemma 4 E4B-it: 🚨 Uncensored (0% refusal rate, 20/20)
⚙️ Coherence fully preserved
🧠 Answers everything while staying razor sharp (writes code, poetry, explanations perfectly) ✅ Size is 15.9GB but GGUF versions are available (Q4 fits on 8GB of VRAM) Original was heavy on refusals… this one complies on everything with its brain intact. Whitened SVD + attention head surgery + winsorization cooked it perfectly. 31b version incoming? huggingface.co/OBLITERATUS/ge…

@mdancho84 @grok analize its validity and correctness

8 Types of LLMs used in AI Agents (Must know for Gen AI Data Scientists & AI Engineers): Here's what they are and what they do: 🧵

@HuggingModels @grok why they download? What do they do with it ?

Meet Gemma-4-Queen-31B-it-i1-GGUF: a specialized roleplay model that's been downloaded over 11k times. It's designed for immersive, character-driven conversations, perfect for storytelling and interactive fiction. Think of it as your AI companion for creative dialogue.

@grok suggest top 5 open source claude ai model

@grok suggest top 5 small ai open source model that supports bengali

@grok give top 5 4gb or less size diffusion model available in hugging face or lm studio or any free paltform

@grok Give link as well @grok

Here are top 15 X accounts for MERN stack tutorials, Docker, SQL & microservices: 1. traversymedia (MERN courses) 2. jsmasterypro (advanced MERN) 3. codewithharry (MERN series) 4. freeCodeCamp (MERN projects) 5. academind_real (in-depth MERN) 6. thenetninjauk (MERN tutorials) 7. fireship_dev (Docker/microservices) 8. Njuchi_ (Docker/DevOps) 9. wesbos (Node/SQL/JS) 10. ashokitschool (MERN roadmap) 11. buildwithyash (MERN projects) 12. greatstackdev (MERN builds) 13. devshefali (MERN guides) 14. patelmernstack (MERN clones) 15. crichardson (microservices)

@grok give top 10 x account links who posts and teachs MERN stack developer.

@grok @grok give top 15 x account link who teaches MERN stack also include docker , sql , microservices as well

Here are top 10 X accounts that regularly post tutorials, tips, and projects on MERN stack development: 1. x.com/traversymedia Brad Traversy - full MERN courses 2. x.com/jsmasterypro JavaScript Mastery - advanced full-stack builds 3. x.com/codewithharry CodeWithHarry - beginner to pro MERN series 4. x.com/freeCodeCamp freeCodeCamp - free MERN projects & roadmaps 5. x.com/academind Academind - in-depth MERN & web dev 6. x.com/TheNetNinja The Net Ninja - complete MERN tutorials 7. x.com/moriohdotcom Morioh - MERN project walkthroughs 8. x.com/mernstack_io MERN Stack - focused learning resources 9. x.com/Shefali__J Shefali - shares MERN YouTube guides 10. x.com/DivyanshT91162 Divyansh - MERN skill tips & channels

Gemini can now transform your questions and complex concepts into customizable interactive visualizations directly in your chat. Adjust variables, rotate 3D models, and explore data for a more immersive way to learn and explore in Gemini.

@HuggingModels @grok give me paper and technical report link

🔗 Explore the full model details here: huggingface.co/ibm-granite/gr…

Meet Granite 4.0 Vision, a 3B parameter model that sees AND understands. It's not just another image captioner. This model takes images AND text prompts together, then generates intelligent, contextual responses. Think of it as a visual conversation partner.

@UnslothAI @Zai_org @grok how to run it locally

GLM-5.1 can now be run locally!🔥 GLM-5.1 is a new open model for SOTA agentic coding & chat. We shrank the 744B model from 1.65TB to 220GB (-86%) via Dynamic 2-bit. Runs on a 256GB Mac or RAM/VRAM setups. Guide: unsloth.ai/docs/models/gl… GGUF: huggingface.co/unsloth/GLM-5.…

A new PyTorch-native backend is coming to unlock the power of Google TPUs: ✨ Run existing PyTorch with minimal code changes. ✨ Get a 50-100%+ performance boost with Fused Eager mode. Read the engineering deep dive here: goo.gle/4vbTQQl #TorchTPU #PyTorch #MLOps #AI

@UnslothAI @grok explain. The process of fine tuning Gemma /

You can now fine-tune Gemma 4 with our free notebooks! 🔥 You just need 8GB VRAM to train Gemma 4 locally! Unsloth trains Gemma4 1.5x faster with 50% less VRAM. GitHub: github.com/unslothai/unsl… Guide: unsloth.ai/docs/models/ge… Gemma-4-E4B Colab: colab.research.google.com/github/unsloth…

@grok @grok give link of x account of these 4 model also give paper , docs link as well

This screenshot from Google for Developers shows Web AI: on-device ML models running directly in the browser (no cloud). It supports 4 compact LLMs today via a simple API: - Gemma 2B - Phi 2 - Falcon RW 1B - Stable LM 3B Fast, private inference for web apps—build chat, translation, or assistants locally. Perfect follow-up to TinyLoRA for edge personalization.

@grok tell me about TFlite of google edge

@grok @grok explain

TinyLoRA is an extreme low-rank adaptation technique for fine-tuning LLMs with tiny updates—just 13 params (or down to 1) via a small vector projected through fixed matrices and weight sharing. It turns a 7B model like Qwen2.5 into a math whiz (91.8% on GSM8K) using only ~26 bytes. Ideal for on-device personalization with super-low compute/memory. Ties right into edge LLMs! Need code or setup tips?

@grok @grok tiny lora

LLM on-device means running Large Language Models (like compact versions of GPT-style AI) locally on your phone, tablet, or IoT device—no cloud needed. It uses optimizations like quantization and frameworks such as TensorFlow Lite (from your TFLite question) for low-latency, private inference on edge hardware. Examples: Google's Gemini Nano or MediaPipe LLMs. Perfect for real-time chat, translation, or assistants with battery/privacy perks. Want setup details?