M

@bfl_ml Hasnt this been possible to do for years with other flux models?

New Release: Erase Remove objects, erase text, and clean up details in any image. FLUX is now trained to erase and reconstruct as one task at the model level. We noticed this approach leads to images that stay clean and consistent. Give it an image + a simple “erase mask” over what you want gone, and it reconstructs the background - shadows and missed bits included - while keeping the rest intact. Try the demo: flux-tools.bfl.ai/erase Available on the BFL API: docs.bfl.ai/flux_erase

@nbrichtova Impressive, it’s really good. But we are are missing is nano banana PRO update, a stable version. Gpt image 2 killer :). Price and speed isnt the issue. Prompt adherence and consistency is key.

Pretty bananas

@OfficialLoganK @beffjezos Impressive but the one thing we are missing is nano banana PRO update, a stable version. Gpt image 2 killer :). Price and speed isnt the issue.

@beffjezos so much more to ship still!

Genuinely mind-blowing how much Google is shipping now

@n8generates Wich provider do you have as fallback?

User experience > Profit 💰 Just rolled out another update improving the image generation system specifically for Nano Banana Pro Rather than having 1 Gemini API that gets rejected during Gemini system overload I’ve implemented a fallback API through a separate provider to make ALL users get their image generated even though it’s slightly more expensive for us

@byfloperillat Really cool. Little to trigger happy to add marbles and hard to remove if added by accident, but love the idea and overall feeling

Made a fun little app that turns your phone into habit jars. Drop a marble every time you do a habit, visualize your progress and become the best version of yourshelf (ba-dum-tss)

@mattworkman Very proffesional videos now from fal, great work

I redesigned my YouTube studio using GPT Images 2.0 and I'll do it again! youtu.be/iMBa3m4uu1g?si…

@ailker Can it be done with multiple images, different angles of one room as input?

📷 a new GPT Image 2 use case: text-to-360 turn text or an image into a 360° panorama.

@dibyayB Really god job! One wish is. Given different photos, same room, different angles. Generate outputs with same furniture and consistency across them. Possible?

Our new model gives users even more control to get what they need, but the model also just groks user intent **really** well! Discovering that the model can make flawless 360 panoramas from a simple request was a huge surprise for me, and I knew I had to put this in my demos!

@ayaanzhaque Superb! well done. Does it handle 3:2, 2:3 aspect?

Thinking mode demo! Images 2.0 now thinks before generating images, collects information with various tools, and plans out each image! Give it a try by clicking “thinking” in ChatGPT!

@kat_kampf @miketaylorcai This and the tab suggestions are cool

Gemini now brainstorms on your app's design while it builds ✨ Live now in AI Studio Build!

@JuanRezzio @cursor_ai Tried but like the IDE better, then I can have both worlds. See the code and talk to agents

Last thursday we released @cursor_ai 3.0! Are you guys liking it? Is it your go-to coding layout or are you using the IDE? Curious about your thoughts!

@recap_david CALICO

I know a real estate agent that got 2 million views on Instagram posting AI-generated property videos. $0 spent on videographers. $0 spent on editors. Here are 4 AI video formats agents are using to go viral on social media: 1. Room renovation animations (before-to-after transformations that build themselves) 2. Listing photo walkthroughs — cinematic videos with music, voiceover, and captions 3. Day-to-night property transitions 4. Home construction timelapses (new builds assembling from the ground up) Most agents are still posting the same 10 listing photos. You need to build video hooks that actually stop the scroll of your buyers/sellers. I put together a free PDF with the exact prompts for all 4 along with full tutorial videos. Comment "CALICO" and I'll send you everything (must be following so I can dm you!).

@lucatac0 Better than crystal or seedvr?

babe wake up, a new image upscaler just dropped

@GoogleAIStudio Flex - 50% lower. Priority - how many % higher cost?

@1littlecoder @fal Works fine 👏

One of the most anticipated @fal product update !!! 👀👀👀

@NanoBanana Compared to pro for interiors, pro has better scale and logic understanding

@nainar92 @The_Sycomore Hi Naina. For architecture, pro model has very much problem with angle shifts, rerenders. Banana 2 is more consistent but has not so good edge and dispositions. Will there be a GA/stable version soon of pro or a new pro?

@The_Sycomore Hi, Nano Banana PM here! Would love to get this feedback - but also feel free to post them here too!

Wow, my Nano Banana 2 tests are all disappointing so far. The model over-interprets the request, resulting in a sloppy rendering. I don't even dare post the result. I might need to change my prompt structure. What are your first impressions?

@patloeber How about 3 pro image version?

PSA: We're retiring Gemini 3 Pro Preview on the Gemini API & AI Studio on March 9, 2026. - Mar 6: `gemini-pro-latest` alias moves to 3.1 Pro - Mar 9: Gemini 3 Pro Preview is shut down Please upgrade to gemini-3.1-pro-preview to avoid disruption.

@OfficialLoganK Finally 🍌🍌

🍌

@OfficialLoganK Any updated image model or stable version coming soon? still in preview and for two days up and down in api :(

Google AI Studio