LLMS

→ View original post on X — @rasbt

Claude 3.7 Model: Clarifying Distillation vs RL Training Scaling

By

@rasbt

–

27 February 2025 22h26

> – Claude 3.7 is the new o3-mini-high
Wait, 3.7 is a distilled model? I thought the system card said it was RL-trained plus it uses inference time scaling?

→ View original post on X — @rasbt

GPT4.5 Technical Details and OpenAI’s Lack of Documentation

By

@rasbt

–

27 February 2025 22h11

GPT2 and GPT3 were the good old times with technical reports. GPT3.5 had at least the InstructGPT report. GPT4 had strong bigger+MoE rumors.
Does anyone know what's new in GPT4.5?

GPT-4.5 Early Testing Shows Natural Interaction and Broader Knowledge

By

–

27 February 2025 22h08

Early testing shows that interacting with GPT-4.5 feels more natural. Its broader knowledge base, improved ability to follow user intent, and greater “EQ” make it useful for tasks like improving writing, programming, and solving practical problems.

GPT-4.5 Advances Pattern Recognition Through Scaled Pre-Training

By

–

27 February 2025 22h08

GPT-4.5 is a step forward in scaling up pre-training and post-training. By scaling unsupervised learning, GPT-4.5 improves its ability to recognize patterns, draw connections, and generate creative insights without reasoning.

GPT-4.5 Features: Search, Uploads, Canvas, No Multimodal Yet

By

–

27 February 2025 22h08

GPT-4.5 has access to the latest up-to-date information with search, supports file & image uploads, and can use canvas to work on writing and code. GPT-4.5 does not currently support multimodal features like Voice Mode, video, and screensharing in ChatGPT. In the future, we

OpenAI releases GPT-4.5 to ChatGPT Pro users

By

–

27 February 2025 22h08

Today we’re releasing a research preview of GPT-4.5—our largest and best model for chat yet.

Rolling out now to all ChatGPT Pro users, followed by Plus and Team users next week, then Enterprise and Edu users the following week. pic.twitter.com/br5win5OEB
— OpenAI (@OpenAI) 27 février 2025

Today we’re releasing a research preview of GPT-4.5—our largest and best model for chat yet. Rolling out now to all ChatGPT Pro users, followed by Plus and Team users next week, then Enterprise and Edu users the following week.

→ View original post on X — @askalphaxiv

Meta’s Llama3-SWE-RL Outperforms GPT-4o on Code Tasks

By

@askalphaxiv

–

27 February 2025 21h56

DeepSeek-R1 proved RL boosts LLM reasoning. Meta now scales it for real-world SWE tasks Llama3-SWE-RL-70B, trained with RL on GitHub issues, PRs, & code fixes Outperforms SFT on math & code reasoning Best medium-sized LLM on SWE-Bench Verified
Comparable to GPT-4o

→ View original post on X — @karpathy

GPT 4.5 Release: Measuring AI Improvement Through Scaling

By

@karpathy

–

27 February 2025 21h42

GPT 4.5 + interactive comparison 🙂 Today marks the release of GPT4.5 by OpenAI. I've been looking forward to this for ~2 years, ever since GPT4 was released, because this release offers a qualitative measurement of the slope of improvement you get out of scaling pretraining