Muriithi Nyaga

CG Specialist with 10+ Yrs of Experience in Motion Graphics, Animated film and MOCAP. I am currently working with locally run LLMs and T2I & I2V Ai models.

[Muriithi Nyaga @TheOneNyaga]

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.

[Muriithi Nyaga @TheOneNyaga]

@HermesWatcher @pablothethinker Just tell him to install Hermes Agent on his Engine Control Module and let it loose YOLO mode. Oh and Don’t forget to add … “Thank me later.” After that 😂

[Muriithi Nyaga @TheOneNyaga]

@TadyJerry Maybe he just needs to change the people around him, his environment has made him think having children fucks up your life.

[MiniMax (official) @MiniMax_AI]

If you're struggling with the inference speed of your local MiniMax H3 model, this is probably what you're missing.🫣

The community loves Sol Engine—thank you for setting so many GPUs on fire and making people feel GPU rich!🩵😝

[Nous Research @NousResearch]

Hermes has twelve browser tools. Browser Use mode replaces them with a single one, driven by @browser_use's CLI 3.0.

Instead of a dozen schemas in every request and a tool call per click, the agent writes a script. In our tests that cut token use 48-66% with no accuracy drop.

[Machine Delusions @Machinedelusion]

https://t.co/7elJCjRh0m

A new / Better distill lora!

[ayesha @ayesha_fatiima]

The meme has evolved.

[⚡AI Search⚡ @aisearchio]

This Minimax H3 workflow is WAY easier. Handles t2v, i2v, r2v all in one simple workflow.

ComfyUI-MiniMaxH3-Easy
https://t.co/IKPW6hIIqK

[Muriithi Nyaga @TheOneNyaga]

The only way to prepare for Qwen 3.8 27b 😂

[Wildminder @wildmindai]

ComfyUI MiniMax-H3 Image Editing.

Just simple node for image-editing and frame interpolation.

https://t.co/15ygAKOMMD

[MiniMax (official) @MiniMax_AI]

Four days after we opened the weights, the community built what's usually a lab deliverable: a distillation LoRA that cuts sampling from 20 steps to 4–8.

This is why we open-sourced. 🫡

[Stable Diffusion Tutorials @SD_Tutorial]

Kijai's Minimax H3
compressed model weight

He achieved this by squeezing the model's weights down to 4-bit (saving massive amounts of space) but running the math in 8-bit (keeping it fast and accurate) using a special layout called AsymW4A8Int8Layout
👇
https://t.co/I4oh4aW1A2

[RyanLee @RyanLeeMiniMax]

We just dropped SOTA open weights for our MiniMax‑H3 video model, and so much happened in 48 hours.

The community ran H3 on untested hardware ($280 gaming GPUs, fully‑offline MacBooks) and built brand‑new unplanned tooling.
Extremely glad we open‑sourced the weights.
Community creations thread 🧵

Hardware requirements are surprisingly modest.
Users validated H3 works within 24h on:
• $280 RTX 3060 (12GB)
• 16‑GB VRAM cards
• RTX 4090 / RTX 5090

A year ago, this‑quality video generation was only available via cloud APIs.
Now a mid‑range gaming PC is your starting bar.

[Unsloth AI @UnslothAI]

Qwen3.8-27B is coming! 🔥

Will run locally on 17GB RAM/VRAM setups.

[Enze Xie @xieenze_jr]

🚀 MiniMax H3, accelerated on Day 1 with Sol Engine!

Within just 4.5 hours, our agent-native Sol Video Inference Engine achieved:
⚡ 3.95× end-to-end speedup over Diffusers
⚡ 2.80× speedup over SGLang
🎬 8× NVIDIA GB200, 1344×768, 24 FPS, 124 frames

The acceleration combines kernel fusion and graph capture, cross-step caching, and training-free sparse attention powered by Sol-Attn—with no distillation, fine-tuning, LoRA, or offline calibration.

We are especially excited to see a powerful open-weight model like MiniMax H3 released to the community. Open models are essential for pushing video generation research, systems optimization, and real-world deployment forward.

We hope this is the beginning of a much more vibrant open-source video generation ecosystem—and Sol Engine will keep working to make the latest models faster and easier to deploy from day one.

🔗https://t.co/XpESoX5OnU

[Stable Diffusion Tutorials @SD_Tutorial]

MiniMax H3 📹🎬Director

A timeline editor in ComfyUI - storyboard prompts, first/last keyframes, image/video/audio references, joint audio, live sampling preview, retakes and shot chaining.

👇
https://t.co/fGGfiq4xRM

[⚡AI Search⚡ @aisearchio]

Here's a fast preview decoder for MiniMax-H3.

It converts the model’s hidden image data into a rough preview while the video is generating, so you can quickly see the progress without waiting for the final result.

https://t.co/iXYxINq7Ji

[Muriithi Nyaga @TheOneNyaga]

I am a simple man, I load X I see new post about minimax performance or quality of life node, I smile, run cmd on my custom nodes folder, comfyui loads, workflow loads, nodes and prompts load, I hit run workflow, 4090 screams, power company smiles, electricity bill comes in, I scream at the bill, pay it with low grumbles, the mood is now right, I fire up X again, I see post about new AI model with 27b parameter at a nice Q4, I smile and the cycle continues. 😂

[Muriithi Nyaga @TheOneNyaga]

@RyanLeeMiniMax 1x1 and 16x9 HD at 1megapixel, 12sec videos took between 10 and 12 minutes to generate on a 128gb ram and RTX 4090

[⚡AI Search⚡ @aisearchio]

This can speed up your Minimax H3 gens by 20-30%

https://t.co/D6cds8ZTtk

Full tutorial coming tonight!

[Okiya Omtatah Okoiti @OkiyaOmtatah]

Kenya’s Ksh 12.8 Trillion debt trap cannot be ignored.71% of our revenues go directly 2 debt servicing,to pay off loans that Kenyans never saw or approved.ANY loan that BYPASSED Parliament is #odiousdebt.If it didn't benefit the people,the people shouldn't pay it back #DeniBandia