Загрузка видео...

Не удалось загрузить видео

На главную

Flux 2 Klein 4B param, dropped at 2 steps with much higher FPS! I also added a couple of LORA's, I'll mess around with. Such a good diffusion model!

developedbyed's profile picture

Dev Ed

71,242 subscribers

109,619 просмотров • 4 месяцев назад •via X (Twitter)

Наука и технологии
LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

🆕 Image super-resolution model just dropped! Superior results even with a single sampling step. 🔥InvSR: Arbitrary-steps Image Super-resolution via Diffusion Inversion.
0:12

🆕 Image super-resolution model just dropped! Superior results even with a single sampling step. 🔥InvSR: Arbitrary-steps Image Super-resolution via Diffusion Inversion.

Gradio

53,331 просмотров • 1 год назад

A visualization of how I think of diffusion in frequency space Diffusion often generates the low frequencies in the earlier steps before generating the higher frequencies in the later steps
0:11

A visualization of how I think of diffusion in frequency space Diffusion often generates the low frequencies in the earlier steps before generating the higher frequencies in the later steps

wh

163,881 просмотров • 1 год назад

My favorite editing model, FLUX.2 [klein] 9B, just got 2x faster: Meet FLUX.2 [klein] 9B-KV 😍💨 > Using KV-Cache Optimization to reduce computation & speed up inference by up to 2.5 times for multi-reference editing love how well it edits "around" the bullets
0:11

My favorite editing model, FLUX.2 [klein] 9B, just got 2x faster: Meet FLUX.2 [klein] 9B-KV 😍💨 > Using KV-Cache Optimization to reduce computation & speed up inference by up to 2.5 times for multi-reference editing love how well it edits "around" the bullets

Linoy Tsaban

30,134 просмотров • 3 месяцев назад

Training FLUX LoRA's on @FAL is simple and you can do it yourself in 5 steps 👇
0:10

Training FLUX LoRA's on @FAL is simple and you can do it yourself in 5 steps 👇

Heather Cooper

35,002 просмотров • 1 год назад

🎨 We're open sourcing 4 new LoRAs trained with new fal FLUX.2 [klein] 4B Trainer! 🖼️ Outpaint - Expand your images beyond borders 🔍 Zoom - Select an area and regenerate it enlarged 🧹 Object Remove - Clean removal of unwanted objects 🎭 Background Remove - Seamless background extraction Model weights are on Hugging Face
0:10

🎨 We're open sourcing 4 new LoRAs trained with new fal FLUX.2 [klein] 4B Trainer! 🖼️ Outpaint - Expand your images beyond borders 🔍 Zoom - Select an area and regenerate it enlarged 🧹 Object Remove - Clean removal of unwanted objects 🎭 Background Remove - Seamless background extraction Model weights are on Hugging Face

fal

29,004 просмотров • 5 месяцев назад

apparently this is the "OS" you get when you combine a 2.2B param RNN "kernel" with a 263M param diffusion-based "renderer"
0:14

apparently this is the "OS" you get when you combine a 2.2B param RNN "kernel" with a 263M param diffusion-based "renderer"

Jack Morris

396,971 просмотров • 5 месяцев назад

The original Nidhogg looked kind of soft and mushy, and the whole model appeared to be a all red mess. So, I added some sense of power and a higher contrast material.😆
0:13

The original Nidhogg looked kind of soft and mushy, and the whole model appeared to be a all red mess. So, I added some sense of power and a higher contrast material.😆

Papachin/怕怕

32,167 просмотров • 1 год назад

I might've had WAY too much fun with oatmeal and a real mess afterwards... Such bulk... 🫠
0:15

Sensitive content

I might've had WAY too much fun with oatmeal and a real mess afterwards... Such bulk... 🫠

Kai!

36,969 просмотров • 2 лет назад

The usual suspect threatens customers at a Naples bar with scissors and gets a good dose of education. In Naples, you don't mess around.
0:17

The usual suspect threatens customers at a Naples bar with scissors and gets a good dose of education. In Naples, you don't mess around.

RadioGenoa

4,598,296 просмотров • 8 месяцев назад

Introducing Jan v3, with updates to Jan Desktop v0.7.6 💛 Jan v3 is our first v3 model, a 4B base built for fine-tuning and fast local use, with stronger math and coding. This release also includes a small Jan Desktop update, starting with a UI refresh as we move toward a more unified Jan experience. Try it: - Jan v3 is available in Jan Desktop and at - Get the latest desktop app at Model: - Jan-v3-4B: - Jan-v3-4B-GGUF: Thanks Qwen for the base model and Georgi Gerganov for llama.cpp 💛
0:17

Introducing Jan v3, with updates to Jan Desktop v0.7.6 💛 Jan v3 is our first v3 model, a 4B base built for fine-tuning and fast local use, with stronger math and coding. This release also includes a small Jan Desktop update, starting with a UI refresh as we move toward a more unified Jan experience. Try it: - Jan v3 is available in Jan Desktop and at - Get the latest desktop app at Model: - Jan-v3-4B: - Jan-v3-4B-GGUF: Thanks Qwen for the base model and Georgi Gerganov for llama.cpp 💛

👋 Jan

53,881 просмотров • 5 месяцев назад

I added a couple easter eggs around the map. :3
0:11

I added a couple easter eggs around the map. :3

Sirhaian

46,492 просмотров • 1 год назад

Last year when Krunal pandya tried to mess with MS Dhoni by bowling a bouncer ❌ MS Dhoni responded with 2 massive sixes of the next 2 balls 😮‍💨☠️ Never mess with MSD i repeat never ever 😎
0:20

Last year when Krunal pandya tried to mess with MS Dhoni by bowling a bouncer ❌ MS Dhoni responded with 2 massive sixes of the next 2 balls 😮‍💨☠️ Never mess with MSD i repeat never ever 😎

DB

152,894 просмотров • 3 месяцев назад

Text-to-image diffusion transformer models learn to align text and image representations as a byproduct of their conditional denoising task. By taking the dot product between the text and image representations of a DiT model (like Flux 2), you can create rich saliency maps.
0:19

Text-to-image diffusion transformer models learn to align text and image representations as a byproduct of their conditional denoising task. By taking the dot product between the text and image representations of a DiT model (like Flux 2), you can create rich saliency maps.

Alec Helbling

94,095 просмотров • 6 месяцев назад

Introducing FLUX.2 [klein]. Blazing fast. Beautiful. Generate stunning images in under a second while maintaining exceptional quality. Great for fast editing, changing styles, and developing ideas from 0 → 1. Available via API, or run it locally - Klein 4B under Apache 2.0, Klein 9B as open weights. Try it for free in our demo app (link in the thread).
0:13

Introducing FLUX.2 [klein]. Blazing fast. Beautiful. Generate stunning images in under a second while maintaining exceptional quality. Great for fast editing, changing styles, and developing ideas from 0 → 1. Available via API, or run it locally - Klein 4B under Apache 2.0, Klein 9B as open weights. Try it for free in our demo app (link in the thread).

Black Forest Labs

445,551 просмотров • 5 месяцев назад

🚨 Real-time image-to-image editing drops on fal! 🎨 10+ FPS for instant results for FLUX.2 Klein ✨ Low latency, production-ready 💰 Hand-tuned kernels for real-time execution 🎬 Image editing at interactive speeds
0:19

🚨 Real-time image-to-image editing drops on fal! 🎨 10+ FPS for instant results for FLUX.2 Klein ✨ Low latency, production-ready 💰 Hand-tuned kernels for real-time execution 🎬 Image editing at interactive speeds

fal

35,880 просмотров • 4 месяцев назад

When you mess up things at BCCI, but you got a dad with no one to mess around with him.
0:13

When you mess up things at BCCI, but you got a dad with no one to mess around with him.

Rofl Gandhi 2.0 🏹 Commentary

44,783 просмотров • 6 месяцев назад

(1/2) Check out "𝐏𝐨𝐥𝐲𝐃𝐢𝐟𝐟: Generating 3D Polygonal Meshes with Diffusion Models"! Our model operates directly on the polygons of 3D meshes and generates novel shapes as output through an iterative diffusion process.
0:11

(1/2) Check out "𝐏𝐨𝐥𝐲𝐃𝐢𝐟𝐟: Generating 3D Polygonal Meshes with Diffusion Models"! Our model operates directly on the polygons of 3D meshes and generates novel shapes as output through an iterative diffusion process.

Matthias Niessner

57,941 просмотров • 2 лет назад

Along with Season 2 we added Impossibles to skate. Here’s a look at just a few of the ways you can do it!
0:14

Along with Season 2 we added Impossibles to skate. Here’s a look at just a few of the ways you can do it!

skate.

22,361 просмотров • 6 месяцев назад

Clarity Upscaler now works with Flux! 🥳 Clarity started 1 year ago as an upscaler to convert images with Stable Diffusion 1.5 into high-resolution, photorealistic images It took awhile to make it work with Flux but now it does. With Flux it's even better at upscaling faces while preserving identity and resemblance (a common problem with other upscalers) LoRAs are also supported, so upscaling with a specific style or face works perfectly Now available on Clarity AI and through my own API Link in the reply!
0:15

Clarity Upscaler now works with Flux! 🥳 Clarity started 1 year ago as an upscaler to convert images with Stable Diffusion 1.5 into high-resolution, photorealistic images It took awhile to make it work with Flux but now it does. With Flux it's even better at upscaling faces while preserving identity and resemblance (a common problem with other upscalers) LoRAs are also supported, so upscaling with a specific style or face works perfectly Now available on Clarity AI and through my own API Link in the reply!

philz1337x

71,216 просмотров • 1 год назад

🐟🐠NEW MODEL REVEAL!!! 🐠🐟 As we set sail for the SEA-son 2, I would much rather show the process than hype up an invisible model for you to see at Season 2! Meet the your long forgotten imaginary friend once again with a new look. With a new raggedy/toylike feel 🧸
0:11

🐟🐠NEW MODEL REVEAL!!! 🐠🐟 As we set sail for the SEA-son 2, I would much rather show the process than hype up an invisible model for you to see at Season 2! Meet the your long forgotten imaginary friend once again with a new look. With a new raggedy/toylike feel 🧸

AlpineShowTime ⭐ Imaginary Friend VTuber ⭐

83,380 просмотров • 6 месяцев назад