OpenAI Launches GPT-5.3-Codex

In partnership with

OpenAI Launches GPT-5.3-Codex, Its Most Capable Coding Agent

OpenAI released GPT-5.3-Codex on February 5, calling it the most capable agentic coding model to date. The model combines frontier coding performance with the reasoning abilities of GPT-5.2, runs 25 percent faster, and can be steered in real time while working on complex tasks. In a notable milestone, OpenAI disclosed that the model was instrumental in creating itself, with engineers using early versions to debug training runs and evaluate performance.

The release intensifies competition with Anthropic, which launched its own coding model just 15 minutes earlier the same day. GPT-5.3-Codex achieved state-of-the-art results on SWE-Bench Pro and Terminal-Bench benchmarks. (Source: OpenAI)

DOWNLOAD FREE Guide: 100 ChatGPT Prompts to Boost Productivity

Want to get the most out of ChatGPT?

ChatGPT is a superpower if you know how to use it correctly.

Discover how HubSpot's guide to AI can elevate both your productivity and creativity to get more things done.

Learn to automate tasks, enhance decision-making, and foster innovation with the power of AI.

Download the free guide

Anthropic Launches Opus 4.6, Its Most Powerful Model Yet

Anthropic unveiled Claude Opus 4.6 on February 5, representing a major leap in enterprise AI capabilities. The new model excels at financial analysis, code review, and long-form document processing, and introduces a breakthrough feature allowing teams of AI agents to collaborate on complex projects in parallel. Perhaps most notably, during internal testing, Opus 4.6 discovered over 500 previously unknown zero-day security vulnerabilities in open-source code libraries with minimal prompting.

This finding signals a potential inflection point for defensive cybersecurity, where AI could become the primary method for securing open-source software. Pricing remains at $5 per million input tokens and $25 per million output tokens.

Free vs Paid AI Tools In 2026

I spent 3 hours reading Reddit threads with 2,000+ comments about ChatGPT prompts

I tested Grok Imagine inside Higgsfield to see if it can recreate real crime-drama tension

AI & Tech News

Perplexity Launches Model Council, Running Multiple Frontier AI Models in Parallel

Perplexity announced Model Council on February 7, a system that runs Claude, GPT-5.2, and Gemini simultaneously to generate cross-validated answers. The approach presents results from different models while noting commonalities and differences, significantly improving reasoning quality and reducing hallucination errors compared to single-model queries.

The development represents a shift from traditional AI query architecture toward what some observers call an "AI committee" that cross-checks each member's work. Industry analysts see this as potentially setting a new standard for AI reliability by leveraging the strengths of different models while compensating for their individual weaknesses through collaborative validation. (Source)

DeepMind CEO Says China's AI Models Now Trail US by Just Months

Google DeepMind CEO Demis Hassabis said Chinese AI models trail U.S. capabilities by just months rather than years, acknowledging that companies like Alibaba and startups including Moonshot AI and Zhipu have closed the gap faster than expected. Speaking on CNBC's Tech Download podcast, the Nobel laureate described working 100-hour weeks for 50 weeks annually in what he called "ferociously competitive" AI development with "incredibly high" stakes.

Hassabis questioned whether Chinese firms can innovate beyond current technological frontiers. On physical robotics, he estimated a breakthrough moment remains 18 to 24 months away, with Google's partnership with Boston Dynamics potentially yielding demonstrations within two years. (Source)

Machina Labs Raises $124 Million to Build AI-Driven Defense Manufacturing

Robotics and advanced manufacturing company Machina Labs closed a $124 million Series C on February 8 to build its first large-scale "Intelligent Factory" for defense and aerospace applications. Toyota's Woven Capital led the round alongside Lockheed Martin Ventures and Strategic Development Fund.

The company integrates forming, machining, welding, and assembly into a single software-defined production system that can be deployed and adapted as dynamically as code. Machina Labs has already secured contracts from the U.S. Air Force Research Laboratory and is working with a major defense prime on metal structures for missiles and hypersonics. The company positions itself as solving a manufacturing bottleneck that limits how quickly modern defense systems can be produced. (Source)

Top AI tools to check out…

I’ve been building a best AI tool directory where you’ll find the best of the best tool and their USECASES.

Thanks for your time!

Shailesh & OpenAILearning team