Build a Large Language Model from scratch! This repository contains the code examples for developing, pretraining, and finetuning a LLM from scratch. It is the official codebase for the book Build a Large Language Model (From Scratch). Notebook examples are included for each
RESEARCH
-
35B AI model tops forecasting leaderboards, rivals larger models
By
–
v. Interesting. A 35B model swept the top of the AI forecasting leaderboards in early June and it's still going toe-to-toe with far bigger ones now Check @Apodex_AI
's Apodex-1.0-mini on FutureX ↓ -
FireworksAI and Alibaba Qwen Partnership for Efficient Traceability Judge
By
–
We partnered with @FireworksAI_HQ to build an efficient traceability judge. We fine-tuned an @Alibaba_Qwen model to detect 'perceived errors' on each production trace. It matched or exceeded the performance of state-of-the-art models and
-
THE PROOF MISSIONS: Public Program for Autonomous Work and Proof-Governed Cooperation
By
–
THE PROOF MISSIONS A public program for turning autonomous work into accepted capability, transfer evidence, safe composition, accountable institutions, and proof-governed cooperation across sovereign institutions. https://
montrealai.github.io/goalos-agialph
a-ascension/proof-missions.html
… #AGIALPHA #ASIFirst -
GLM-5.2 open-weight model with 1M context window released
By
–
/3 http://
Z.ai drops GLM-5.2 weights on Hugging Face. GLM-5.2 is a flagship open-weight model built for long-horizon tasks, especially coding and agentic work. Its biggest headline is a stable 1M-token context window, giving it room to handle large codebases and -
Waymo Sensor Fusion Part 1: LiDAR, Radar, Cameras
By
–
(How Waymo Sees — Part 1)
— Satya Mallick (@LearnOpenCV) 23 juin 2026
LiDAR tells a Waymo where everything is. Radar tells it how fast everything's moving. Cameras add color and texture. No single sensor is enough — autonomy lives in the fusion.
Full deep-dive 👇https://t.co/axILFYfFQW pic.twitter.com/bpPFiO7TD5(How Waymo Sees — Part 1)
LiDAR tells a Waymo where everything is. Radar tells it how fast everything's moving. Cameras add color and texture. No single sensor is enough — autonomy lives in the fusion.
Full deep-dive https://
learnopencv.com/3d-lidar-visua
lization/
… -
Chinese models efficiency with MoE and open source vs proprietary AI
By
–
Is there also not a driver in Chinese models have more efficiency w/ MOE and other innovations in open source stack vs proprietary AI models? And also maybe because US players are providing data centres to most of the world?
-
ASI emerges 3 months post-AGI but not 3 months from now
By
–
It may be possible in some scenarios that ASI occurs 3 months after AGI but ASI is not 3 months away from today.
