Popular generative ai tools
-
Dan Dascalescu🙏 75 karmaNov 11, 2024@GammaMakes it easy to get started with a presentation if you have an outline, but as soon as you want more control, you're better off using Google Slides. Gamma's feature that I used the most was image generation - you can select among a variety of models (Ideogram, Flux, DALL-E, but not Midjourney), and it shows 3 variations to choose from. Performance is slow and a little janky, even in Chrome. Prompts and settings are often not respected - e.g. you set it to "preserve text" and just generate slides, and it goes onto dumping an entire 3 paragraphs of text in one slide. You can't overlay text over images (e.g. for image attribution), and there's no precise positioning control, or grouping elements. No way to control table layout, e.g. to have two images slide-by-side in full bleed mode. I tried Gamma for a new presentation, but next time I need to make a presentation, I'll go back to Google Slides and generate images independently. -

-

-

-
Very good, they give a lot of loans, unlike other similar ones, and the quality is good and convenient -
I got some top quality results after learning how to prompt songs 🤩 -
too expensive for me, I just want to make memes, not pay that much -

-

-

-
Classic tool 😮💨 Loving Gen 4, it's wild how good the motion consistency is. It feels like a full studio in your browser. Big W guys keep it up! -
quick and the quality's nuts. tons of presets and styles to pick from, definitely one of those tools that makes you wanna keep testing more stuff. -

-
Diffusion pulls out extremely good images from simple prompts. Renders were a little slow on higher quality, but totally worth it for the results. The controls feel clear enough and the outputs look legit. Good one here! -

-

-

-
Claude — v4.8Claude Opus 4.8 Best in class coding: record 69.2% on SWE-Bench Pro, up from 4.7's 64.3% New fast mode: ~2.5x speed, about 3x cheaper User-controlled effort levels (high default, plus "extra"/"max") Dynamic Workflows (research preview): spins up hundreds of parallel subagents in a single session More honest: roughly four times less likely to let coding errors pass unremarked -

