1. there is no fork of pytorch fwiw. only a fork of nccl. clearly you overheard wrong.
2. torch.compile wasn't even considered because it doesnt have precompilation yet and that was deemed a requirement for large-scale xlformers runs. so the whole thing about graph breaks is
LLMS
-
PyTorch Compilation Strategy for Large-Scale Transformer Models
By
–
-

GPT-5 Visual Reasoning Capabilities Show Weakness in Testing
By
–
gpt5 visual reasoning is the weakest so far in my testing
-
Google’s 270M-parameter Gemma targets Apple market
By
–
The 270M-parameter Gemma is Google's pitch to Apple.
-
100+ Free AI Agent Tutorials: RAG, Voice, MCP Systems
By
–
100+ free step-by-step tutorials with code covering: AI Agents RAG Systems Voice AI Agents MCP AI Agents Multi-agent Teams Autonomous Game Playing Agents P.S: Don't forget to subscribe for FREE to access future tutorials.
-
100+ Free Step-by-Step AI Agents and RAG Tutorials
By
–
100+ free step-by-step tutorials with code covering: AI Agents RAG Systems Voice AI Agents MCP AI Agents Multi-agent Teams Autonomous Game Playing Agents P.S: Don't forget to subscribe for FREE to access future tutorials.
-
LLM Security: Mixing Developer Instructions with Untrusted Input
By
–
Pretty much inherent to how we build software on top of LLMs – mixing developer instructions with untrusted user input is an anti-pattern that's baked into how we build this stuff
-

ChatGPT can now access Gmail and Calendar
By
–


ICYMI: Now you can prompt ChatGPT to access gmail and Google Calendar and it will open an option to connect with these apps. Tool use behind GPT-5 is getting to a new level. Seems like it is now rolled out broadly.
-
Weekend Discussion: Top Use Cases for Low TTFT
By
–
It’s the weekend. Let’s talk low TTFT. What are your top use cases?
-

AI Services Specializing Between Persuasiveness and Accuracy
By
–
The future is hard to predict. Still, one read on "AI persuasiveness decreases AI accuracy" is that AI services could speciate on prioritizing "warmth" (sycophancy, consumer capture) or accuracy. Personality-first: ChatGPT, Meta, Grok.
Accuracy-first: Gemini? Claude API?