AI News: Gemini 2.5 Flash, o3 and o4, Claude Research, Kling 2.0, and More!
# AI roundup: Gemini 2.5 Flash leads wave of powerful new models
This week has been a whirlwind of AI model releases from the major players, with each company trying to outdo the others in capabilities, efficiency, and pricing. Let’s break down the highlights.
## Gemini 2.5 Flash: Google’s budget-friendly powerhouse
Google has released Gemini 2.5 Flash, a more affordable version of their impressive Gemini 2.5 Pro. The standout feature? It’s their first “fully hybrid reasoning model” that lets developers turn “thinking” on or off depending on the task complexity.
What makes it special:
– Super affordable at just $0.15 per million input tokens
– Option to set a “thinking budget” with fixed tokens
– Performs well on benchmarks, though not quite at the level of OpenAI’s new O4 Mini
For context, here’s how the pricing compares:
– Gemini 2.5 Flash: $0.15 per million input tokens
– Claude 3.7 Sonnet: $3.00
– GPT-4o Mini: $1.00
– DeepSeek R1: $0.55
## OpenAI’s triple release: O3, O4 Mini, and GPT-4.1
OpenAI wasn’t playing around this week, dropping three new models:
**O3**: Features the most impressive tool use capabilities seen yet, allowing it to use tools within its chain of thought process. In a striking demo, it was able to pinpoint an exact location in Hawaii just from a photo with no metadata.
**O4 Mini**: A smaller, more efficient, and less expensive version that still performs incredibly well on benchmarks.
**GPT-4.1**: The successor to GPT-4o comes in three versions (Nano, Mini, and full version), offering improved multilingual understanding and reduced latency.
## Anthropic adds Google Workspace integration
Claude has launched “Research” (similar to deep research features from competitors) and added integration with Google Workspace products including Gmail, Calendar, and Docs. This means you can now use AI to draft email responses and create content directly within these tools.
##
Recent Videos
Hermes Agent Master Class
https://www.youtube.com/watch?v=R3YOGfTBcQg Welcome to the Hermes Agent Master Class — an 11-episode series taking you from zero to fully leveraging every feature of Nous Research's open-source agent. In this first episode, we install Hermes from scratch on a brand new machine with no prior skills or memory, walk through full configuration with OpenRouter, tour the most important CLI and slash commands, and run our first real task: a competitor research report on a custom children's book AI business idea. Every future episode will build on this fresh install so you can see the compounding value of the agent in real time....
Apr 29, 2026Andrej Karpathy – Outsource your thinking, but you can’t outsource your understanding
https://www.youtube.com/watch?v=96jN2OCOfLs Here's what Andrej Karpathy just figured out that everyone else is still dancing around: we're not in an era of "better models." We're in a different era of computing altogether. And the difference between understanding that and not understanding it is the difference between being a vibe coder and being an agentic engineer. Last October, Karpathy had a realization. AI didn't stop being ChatGPT-adjacent. It fundamentally shifted. Agentic coherent workflows started to actually work. And he's spent the last three months living in side projects, VB coding, exploring what's actually possible. What he found is a framework that explains...
Mar 30, 2026Andrej Karpathy on the Decade of Agents, the Limits of RL, and Why Education Is His Next Mission
A summary of key takeaways from Andrej Karpathy's conversation with Dwarkesh Patel In a wide-ranging conversation with Dwarkesh Patel, Andrej Karpathy — former head of AI at Tesla, founding member of OpenAI, and creator of some of the most popular AI educational content on the internet — shared his views on where AI is headed, what's still broken, and why he's now pouring his energy into education. Here are the key takeaways. "It's the Decade of Agents, Not the Year of Agents" Karpathy's now-famous quote is a direct pushback on industry hype. Early agents like Claude Code and Codex are...