Muriithi Nyaga

The Work of The Concept Artist, The Storyboard Artist, The Voice Actor, The Modeling Pro, The Environment TD, The Rigging TD, The Physics and Simulation TD, The Foley and Scoring Master and The Animation Guru Starts After The ScreenWriter Has Punctuated To End The Screenplay.

like this LTX2.3 CrossView LoRA. You're basically getting a multi-cam setup from a single raw clip. video + a camera angle prompt, get the same scene re-rendered from a completely different viewpoint. Subject stays, camera moves. huggingface.co/Cseti/LTX2.3-2…

LTX2.3 Multiple-Subject-Reference V2 LoRA. character identity and outfits actually stay consistent now + the annoying flickering is mostly gone. feed it up to 5 refs at once and it blends them naturally. Character movement and scene interaction finally make sense huggingface.co/LiconStudio/LT…

Krea 2 😊Identity Edit V1.1 Compared with V1 its Improved face likeness and image fidelity, much stronger edit (camera, pose, and untouched elements stay fixed far more reliably), better 2-person identity separation, more reliable object remove/replace 👇 huggingface.co/conradlocke/kr…

🤗 MOSS-VL-Realtime is now open source on @huggingface . Built for real-time visual understanding over continuous video streams: 🧠 11B vision-language model 📜 Apache-2.0 license 💬 Ask questions at any point in a video stream 👀 Keeps watching while generating a response 🔄 Revises or interrupts its response as the scene changes 🤫 Can remain silent when more evidence is needed 🧩 256K-token context window 🌐 Chinese and English multimodal understanding 📦 Base, Instruct, and Realtime models are all open source From “watch first, answer later” to “keep watching while answering.” 👀 @Open_MOSS Thank you @sgl_project @lmsysorg for day-0 support! 🚀

@alex_prompter Pray for me

I spend more time tracking my usage limits than actually using Claude. Weekly credits. Sonnet limits. Opus limits. Fable credits. 5-hour windows. Temporary increases that last "a few more days." The rules change every other week. I've been building AI workflows for 50+ clients. Claude is the best model I've used. And being a paying customer has become genuinely stressful. Not because the model is bad. The model is incredible. Because I'm constantly doing mental math on whether I can finish my work before I hit a wall. It's like watching your phone at 8% during an important call. That low-battery anxiety. Every single day. "Compute is expensive." Yeah. We know. But that's your engineering and pricing problem. Not something you convert into daily stress for the people paying you. Right now, Claude Pro feels less like a subscription and more like a rationing system. We don't want credits we need a calculator to track. We don't want limits that change every Tuesday. We don't want "temporary increases." We want to pay, open the app, and work. That's the whole ask. If AI companies don't solve this now, the entire industry is heading somewhere ugly: users spending more cognitive energy managing AI quotas than using AI to work. Fix it while it's early @AnthropicAI.

Meet the LTX 2.3 Storyboard playground 🎬 LTX folks made a great point about text-to-video models use case, so I turned their practical guide into a ZeroGPU space for fast, cheap iterations before full video gen > iterate on your storyboard quickly > easy to customize (duplicate the space and build on top for free) > it's really fun try it: huggingface.co/spaces/ltx-com…

I just converted the LTX 2.3 model to int8-convrot format that is 50% smaller file and now the gen time was just 140.48 seconds. That's 2min faster than the previous one. You will be the judge on the quality.

@milliemilkster @TrollFootball Admittedly they do 😂

@TrollFootball The screenshots before the upset always hit different 💀

Argentina vs Cabo Verde

@signulll You sit on your couch all day so your compression of how much of an advantage this is to an athlete is incompressible to a slob like you.

this is ridiculous lol.

@Bossyator The saddest image is that of the child looking down on the RIP DAD wreath.

Yet to believe it man. I had to take a break from the internet for a week.

@FishAudio Does it have VC Voice clone, ref voice to source voice?

Our best voice model is now free for developers. S2.1 Pro- the same TTS on our paid tier, free API, 83 languages, no hard usage cap, same endpoint you already call. Already integrated? Set model: "s2.1-pro-free" and you're on S2.1 Pro.

Hermes Agent v0.18.0 - The Judgement Release Changelog below:

⏰ Local MCP server for ComfyUI. Inspect workflows, edit graphs, queue generations, and manage models from MCP-compatible agents. This is a very powerful MCP server exposing virtually everything about comfy to the agent! Forming a bridge that allows your agent access to both the front end and backend systems. of ComfyUI What it can do: • Inspect the current open workflow • Read node titles, inputs, outputs, links, and positions • Edit graphs and reconnect nodes • Compact and organize messy workflows • Queue generations from an agent • Discover local models, LoRAs, VAEs, and checkpoints • Debug missing models or broken node links • Take screenshots of the live ComfyUI canvas • Work through the browser bridge or ComfyUI API The agent is even able to code node packs for you, fully manage the running instance of comfy, as well as Verify with screenshot integration of graph state. Iv been sitting on this one for a long while, its time to see some crazy stuff you guys do with it!

@shawmakesmagic Yoh ☠️☠️☠️ shut TFU

They don't even have to ban open source They just have to ban Huggingface and it's over

Introducing Mixture of Agents 2.0 in Hermes Agent. Combine any provider's models into a mixture of your own. Access your presets as if it were a normal model in Hermes. Big improvement in our soon-to-release HermesBench against opus and gpt-5.5 with MoA using Opus & GPT together.

Well… my last GPU just bit the dust. Spent the day trying everything to rule out every other possibility. I guess I’m taking a forced vacation until I can figure something else out. 🤬

@NelsonAkoth_ @kijana_misa @__MadRojas Tell them siku hizi tunafanya home delivery 😂😎

@kijana_misa @__MadRojas These ones came to use men for free wakapata majamaa wamechanuka. How the hell are you coming to drink of my Martell, Hennessy nkl and get back to your house untouched? Who does that?

Netizens are losing their minds after a furious clubgoer exposed a group of women who allegedly camped at Al-Faker all night. Dressed in expensive outfits, they reportedly only ordered a single shisha and one soda. Now women are in the group chats asking, “Where did all the generous men disappear to?”

LTX‑2.3 3D Real IC‑LoRA Turns a rough grey 3D viewport animation (a simple low‑poly blockout) into a fully photorealistic, film‑quality video 👇 huggingface.co/fal/LTX-2.3-3D…

The strongest models are gated and access is granted only to a select few. Hermes Agent now exposes MoA presets as virtual models, giving you capabilities beyond the publicly available frontier: 8% higher than Opus 4.8 and 11% higher than GPT 5.5 on our upcoming benchmark.

Really happy to share this one 🙌 I trained this render-to-real IC-LoRA for LTX-2.3 on 3D data I made myself + a few synthetic scenes. It's the first LoRA of its kind, so still a bit prototypal — it shines on a single scene, and for multiple scenes it's best to generate them separately. Biggest tip: nail your first frame. The reference first frame should closely match the opening frame of your video — that's what makes the result clean and coherent. Hope you all get great results with it ✨ Run on fal: fal.ai/models/fal-ai/… Model on Hugging Face: huggingface.co/fal/LTX-2.3-3D…