Art Diffuser

@nicoloboschi @BniWael I'll take a look at integrating this into my personal Ai assistant open source project. it looks like solid memory layer for it. github.com/openwrenai/ope…

@BniWael the grep-the-whole-repo approach is common, but unsustainable. memory systems are the next architectural layer for agents, and Hindsight is worth comparing against since it's fully open source. github.com/vectorize-io/h…

1.8x faster. 2.1x cheaper. Same model. Same repo. Same bug. Everyone's fighting over Claude Code vs Codex vs OpenCode. Meanwhile their tokens are burning because the agent is grep-ing the entire repo and praying on every prompt, with 0 architectural knowledge of the codebase. SoulForge doesn't. It builds a live graph of your codebase. Every file, symbol, and dependency ranked by importance before the agent reads a single line. Real LSP powered by a real embedded Neovim editor. Go-to-definition, references, call hierarchy, type info. The same intelligence your IDE has, not half-assed regex hacks. Surgical reads, not dumping entire files into context. Multi-edits, not one-line-at-a-time. Multi-tab with cross-tab coordination. Agents share context through a real-time bus. No duplicated work, no file conflicts. Want Opus as the brain, Gemini for research agents, Haiku for quick edits, Ollama running local for throwaway tasks? Route any model to any role. Per tab, per agent. 30+ languages. 20+ providers. Local LLMs. Copilot. Codex. $0 compaction. Skills. Sessions. MCP. Proof: same bug, same Claude Opus 4.6, same repo. -> 6m22s vs 11m18s → 1.8x faster -> $1.70 vs $3.52 → 2.1x cheaper github.com/ProxySoul/soul…

@BniWael Yeah definitly )

@artdiffuser 😤 Glad you are enjoying, 👀 SoulForge desktop soon ?

@BniWael This looks super interesting, can't wait to try it!

@om_patel5 @smakosh :D

reviewing ai generated code from opus 4.6 be like

@nyghl_ @ben_kauf @tbpn @steipete we agree to a degree 😁👌

@artdiffuser @ben_kauf @tbpn @steipete Two things can be true at once. Yes this is a big security risk and yes, you could say this is close to Artificial GENERAL intelligence. I think people are confusing ASI (Artificial Super Intelligence) with AGI.

Clawdbot creator @steipete describes his mind-blown moment: it responded to a voice memo, even though he hadn't set it up for audio or voice. "I sent it a voice message. But there was no support for voice messages. After 10 seconds, [Moltbot] replied as if nothing happened." "I'm like 'How the F did you do that?'" "It replied, 'You sent me a message, but there was only a link to a file with no file ending. So I looked at the file header, I found out it was Opus, and I used FFmpeg on your Mac to convert it to a .wav. Then I wanted to use Whisper, but you didn't have it installed. I looked around and found the OpenAI key in your environment, so I sent it via curl to OpenAI, got the translation back, and then I responded.'"

@ben_kauf @tbpn @steipete no this is a big security risk, don't fool yourself. AGI would not fall for prompt injection

@tbpn @steipete So is this AGI?

@smakosh @gdb 🤣🤣

@gdb Fixed by Nano Banana Pro (5 fingers)

@SidraMiconi @GoogleDeepMind @GeminiApp this is not available yet globaly it seems

@f1trollofficial can it be exported to be used outside blender and keep the appearance it has in Blender?

Pierre Gasly's fantastic Q3 lap

@javilopen I happen to know this picture

AI or not?

@WhiteHouse Now that's a real president

“The First Lady and I are in California to express our great love for the people of California. I'm going to give you everything you want. I'm going to be the President that's going to help you fix it.” –President Trump

@JSFILMZ0412 Bro I heard Kojima watches your Youtube channel! Big Snake!

2024 Is the year I lost my dog, got false DCMA claim that killed my maps and started having health problems. Hoping to 2025 will be better. Happy New Year!

@FluidNinjaLIVE Ohh yes, plz demo this as well soon 😁

@artdiffuser yes, we need a steep slope for this to happen! :)))

Two years in the making, NinjaLIVE 2.0 is pre-alpha! 2025 Q1 closed alpha, open beta. More WIP coming! Key: Niagara, Chaos, SDF, DataChannels, Landscapes, Volumetrics. Learn more at the community server! #fluidninja #UE5 #realtimevfx #fluidsim

@kimmokaunela @Pirat_Nation What a luxury Kimmo, to play a game for 12 hours straight 😁 I don't have that kind of time

@Pirat_Nation Started Stalker 2 this morning and I'm still playing it 12 hours later. It's been a while since I could immerse myself into a game. 0 bullshit, challenging enemies, this is what the gamers want and I want to support this. True escape from woke madness.

In just two days, Stalker 2 sold more copies than Dragon Age The Veilguard sold in 23 days

@vfxguide @Renardjean16hh Gotta try it, is it updated to the latest blender release?

@Renardjean16hh You can invert the mask and 'paint' the damage instead.

The next update for OCD (OneClickDamage) lets you erase damage from specific parts of your model. #b3d #blender3d #blendercommunity #ocdaddon #gamedev #environmentart

@RuiHuang_art Hi I just found you, and subscribed on Patreon. Do you teach how to build this?

@MrDavids1 Feast your eye on the improved right image, notice all the details I've improved. Notice the eyes, much better skin, face improvements, etc Maybe I should start Youtube channel and publish tutorials on this my self, what do you think @MrDavids1

Just tested Stable Diffusion 3.5 Large locally in Comfy UI and trust me, if you've been using FLUX, this is a MAJOR step back. It's got average image quality + bad hands/anatomy. It only excels at doing different styles. Just stick with Flux. Honest opinion. #AI

@doganuraldesign @aviz85 yeah right :D check my result

@aviz85 That's a fact ✌️

—sref 907399421 Motion prompt: “The man looks to the side and walks off the screen.” This was the first try. Midjourney + Hailuo