Daily AI @DailyAIstuff

Claude Opus 4.5 is AGI. The hard part is done We just haven't built the tooling on top of it yet to make it sentient This is where the multi-trillion dollar opportunity is in 2026. Literally anyone can do this (including you)

@dr_cintas Wow I will try now

@dr_cintas Gold

If you want to master AI in 2026, this might be the best video to start with. Karpathy released a free 3.5-hour deep dive on how LLMs like ChatGPT actually work.

Google really cooked with Gemini 3 Pro it just generated me this 3d interactive balloon shooting game in Emergent with a single prompt uses the camera to detect different hand motions, has difficulty levels, sensitivity and smoothing settings tutorial and prompt in comments

What a wild week in AI 🤯 - Gemini 3 Flash - OpenAI’s Image Model - GetStream Vision Agents - Wan 2.6 Multimodal Video - Meta Sam Audio Separation - Microsoft Trellis 2 Image to 3D - Nvidia Nemotron 3 Open Models Here’s EVERYTHING you need to know:

@dr_cintas I was literally working on a project that needed this

@dr_cintas So much happening

What a crazy week in AI 🤯 - OpenAI GPT-5.2 - Mistral Devstral 2 - Cursor Visual Editor - Odyssey-2-Pro Model - Google Disco & GenTabs - Runway GWM World Model - Zai Open-Sources AutoGLM Here’s EVERYTHING you need to know in 1 minute:

Figma just launched AI image editing tool.. to replace Photoshop powered by Gemini 3 with Nano Banana its so over for traditional softwares

camera shot and angle reference chart for your subject with nbp MCU = Macro Close Up MS = Medium Shot OS = Over the Shoulder WS = Wide Shot HA = High Angle LA = Low Angle P = Profile ThreeQ = Three-Quarter View B = Back View prompt (+ ref image): {   "project_name": "Auto_Cinematic_9_Angle_Grid_Generator",   "version": "3.0 (Angle & Anatomy Focus)",   "instructions_for_ai": {     "step_1_analysis": "Analyze the input image for subject identity, lighting (e.g., prism effects, direction), skin texture, emotion, and color palette.",     "step_2_inference": "If the input is a close-up, you must logically infer the subject's outfit, body type, and environment based on the style of the face. Maintain strictly consistent character design across all 9 panels.",     "step_3_execution": "Generate a 3x3 grid where each panel corresponds to the specific camera definitions below."   },   "camera_angle_specifications": {     "MCU": "Macro Close Up: Focus intensely on facial details, eyes, or textures. Crop top of head and chin.",     "MS": "Medium Shot: Waist or chest up. Standard cinematic portrait framing.",     "OS": "Over the Shoulder: Camera placed behind a vague foreground element/shoulder, looking at the subject.",     "WS": "Wide Shot: Full body shot. Show the subject's posture, outfit, and relationship with the environment.",     "HA": "High Angle: Camera is physically higher than the subject, looking down. Emphasize vulnerability or diminishing size.",     "LA": "Low Angle: Camera is physically lower than the subject, looking up. Emphasize dominance or stature.",     "P": "Profile: Strictly from the side (90 degrees). Subject looks completely left or right.",     "ThreeQ": "3/4 View: Subject turned 45 degrees away from the camera. Classic portrait angle.",     "B": "Back View: Camera is directly behind the subject. Seeing the back of the head/body."   },   "output_format": {     "grid_layout": "3x3",     "aspect_ratio": "16:9",     "labeling": "Must include white text abbreviations (MCU, MS, etc.) in the top-left corner of each panel."   },   "final_prompt_instruction": "Using the provided input image as the absolute ground truth for the character and style, generate a photorealistic 3x3 grid collage. You must strictly adhere to the 'camera_angle_specifications' defined above for each panel. Ensure distinct visual variety between the shots (e.g., a Wide Shot must look significantly different from a Close Up). The lighting and color grading must remain identical to the input source in every single angle.\n\nGrid Order:\nRow 1: MCU, MS, OS\nRow 2: WS, HA, LA\nRow 3: P, ThreeQ, B" }

Also, we have a few little Christmas presents for you next week!

GPT-5.2 just dropped and if it wasn’t clear before: There is no wall in AI.

ARC Prize @arcprize

7 months ago

A year ago, we verified a preview of an unreleased version of @OpenAI o3 (High) that scored 88% on ARC-AGI-1 at est. $4.5k/task Today, we’ve verified a new GPT-5.2 Pro (X-High) SOTA score of 90.5% at $11.64/task This represents a ~390X efficiency improvement in one year

156 662 5K 2.3M 1K

Nano Banana Pro with video is insane. The 4k really makes it mind-blowing (made with Hailuo AI)

This AI agent lets you edit Nano Banana Pro images with high precision. You just need to point to the exact detail you want to change (clothes, text, objects), mark the area, and describe the changes.

It’s only been 3 days and we already have: - DeepSeek new v3.2 Models - Runway Gen 4.5 - OpenAI Red Code - Mistral 3 Open Source - ByteDance Seedream 4.5 Image Model - Kling O1 and 2.6 with Audio - Amazon Nova 2.0 and Agents December is going to be insane for AI.

There is no wall in AI. DALL•E Mini vs Nano Banana Pro

google really played the long game. > nano banana 3.0 is insane.

keyboard_arrow_left Previous keyboard_arrow_right Next

Sundar Pichai @sundarpichai

7 months ago

You went 🍌🍌 for Nano Banana. Now, meet Nano Banana Pro.  It’s SOTA for image generation + editing with more advanced world knowledge, text rendering, precision + controls. Built on Gemini 3, it’s really good at complex infographics - much like how engineers see the world:)

814 2K 24K 4.5M 5K

@dr_cintas So good

nano banana + seedream 4.0 + veo 3.1 for fully style consistent videos pipeline and prompts are in the thread below