Riverflow isn’t a typical image model. It actually understands your intent, so even complex edits like text swaps, color changes, or fine detail tweaks come out exactly how you meant them.

On the @ArtificialAnlys Image Editing leaderboard, Riverflow ranks at the top, outperforming every other editing model on most tasks. It doesn’t guess. It designs.

Most image models can’t follow precise edit instructions - they guess or just don’t do anything. Riverflow doesn’t. It understands what you mean and gets it right the first time. Prompt: “Add another dog.” 🐶 Riverflow → adds a realistic third dog seamlessly 🍌 Nano Banana →

Riverflow 1: เครื่องมือจาก Sourceful ที่กำลังเขียนกฎใหม่ของการออกแบบผลิตภัณฑ์ มันไม่ได้เพิ่งเกิดขึ้น แต่ตอนนี้มันเริ่มใหญ่เกินจะมองข้ามได้แล้ว Riverflow 1 จาก Sourceful คือโมเดล AI

A new SOTA image-editing model is here! It delivers incredible results with chain-of-thought reasoning. If you're a professional designer or app builder, try Riverflow 1 on @runware. Here are five great examples ⤵️ 1. From a top-down view to various angles

For Sourceful users, Riverflow 1 is now our default model for in-platform edits including variations, mockups, angles and text changes. That means you can create packaging that feels like you just hired the best agency: 🎨 Text edits that stay perfectly clean 📦 Color changes

Riverflow 2 Preview from Sourceful takes #1 in the Artificial Analysis Image Editing Leaderboard for All Listings, introducing a unified text-to-image and image editing pipeline using Sourceful's reasoning model combined with open weights diffusion models and Seedream 4.0

Riverflow 1 from Sourceful debuts at #1 in the Artificial Analysis Image Editing Leaderboard for All Listings, using a vision language model trained by Sourceful to allow chain of thought reasoning in combination with a third-party open weights diffusion model Riverflow 1

Riverflow 2 is coming later this year but we have more stuff to share before we get there...

🏹❄️ vs 🤓 by Riverflow 1.1 Pro on @runware.

Bring your image to life using Start + End Frame! Seedream 4.0 ➡️ Riverflow 1.1 ➡️ Seedance 1.0 Pro Full process below 👇

It’s called Riverflow, and it was built by Sourceful.

🍀 》I finished naming the shuffle units... It took me lots of time, ue You guys may use this in your character profiles now!! :D Anemones - DOUBLE/NOTHING "Double or Nothing" Sunflowers - SUN+FUN "SUN FUN" Roses - L♡VE 4 ME "LOVE FOR ME" Hydrangeas - RIVERFLOW

AI画像編集のゲームチェンジャー爆誕！🚀「Riverflow 1」が世界No.1に！ ✎. FYIG: https://t.co/9sThh6WGRn 面倒な画像編集作業、もう終わりにしませんか？🤔 Sourcefulが開発した新AIモデル「Riverflow 1」が、数々の強豪を抑えImage Arenaで世界トップに躍り出ましたね！✨ 💡何がすごい？ ✅

Original image - Riverflow 1 - Seedream 4.0 - Nano-banana 🍌 "Make the water pink"

Here are some early opinions from me on the @Sourceful Riverflow 1 model, the current state-of-the-art image editing model. - The model performs better than Nano Banana on most prompts, especially the complex ones. - The resolution is somewhat too low... 🧵

🚨 A new thinking image editing model, "Riverflow," just dropped on Artificial Analysis! It's a total game-changer, feels smarter than nano-banana. 🤯 Is this a OpenAI secret image upgrade or a new beast from Google?

Riverflow 1 tops the Artificial Analysis Image Editing Leaderboard It's around twice as expensive as nano-banana or Seedream 4.0, but it's the first model from Sourceful as far as I know, and they're already working on Riverflow 2!

Riverflow, a new anonymous image editing model, is now on Artificial Analysis It does feel much better than nano-banana, smarter, maybe it's a new version of it or a completely new model, from Google or other company