@patch_gg

Build and play AI mods and games

 San Francisco
discord.gg/patchgg
Joined October 2026
  • Tweets50
  • Following61
  • Followers2.1K
23Media
Include

Only show posts containing:

Exclude

Hide posts containing:

Time range
-
Minimum likes
> Build an 18th-century Brown Bess flintlock musket from scratch in three.js DeepSeek V4.1 Flash: 91 parts Kimi K3: 51 parts Claude Opus 5.5: 65 parts
patch
@patch_ggOct 26
Forged in Fire AI models edition > Make an artisanal 1800s pistol using Blender MCP Claude Opus 5.5: $1.62 GPT-6 Astra: $2.58 Gemini 4 Argon: $1.03 Kimi K3: $0.75
5
221
How to get entire 3D worlds in Blender from a single image
1
3
14
642
okay I got batching working on Nvidia's Motionbricks, and it's still realtime! About 10k generations a second on one 4090, 26x faster than one at a time. You can now have insanely dynamic crowds in your games! opensource: github.com/patch-gg/crowdbri…
Alvaro Bartolome
@alvarobarttOct 26
💚 NVIDIA has released the MotionBricks weights on the Hugging Face Hub! MotionBricks generates real-time motion for game characters and humanoid robots. Give it a character's recent poses and the poses it should reach next, and it generates the movement in between, at 15,000 FPS with 2 ms latency. - One model trained on 350K+ motion clips - Smart Primitives for navigation and object interaction, no retraining - SOMA weights for human characters - Unitree G1 weights for the humanoid robot - Code under Apache 2.0, weights under the NVIDIA Open Model Agreement Animation and robotics motion generation, now open for anyone to build on!
10
20
3
205
6,416
I may have cracked how to animate non humanoid assets with AI! H3 Max generates a video from front and side then we use a finetuned DINOv2 model to track joints! The vid explains the whole process Open source: github.com/patch-gg/anymove
45
100
12
1,050
55,958
Introducing Boss Babies vs Toddlers A FPS game you can play in your browser boss-babies-vs-toddlers.verc… @threejs #threejs #indiegame
5
6
15
603
Which 3D diffusion model should you use for your game assets? We compared the 4 leading models side by side.
8
4
46
3,564
Forged in Fire AI models edition > Make an artisanal 1800s pistol using Blender MCP Claude Opus 5.5: $1.62 GPT-6 Astra: $2.58 Gemini 4 Argon: $1.03 Kimi K3: $0.75
10
1
66
4,585
We've been experimenting with new Blender skills and the results are honestly very impressive
2
4
2
31
2,247
This is insane! I got opus 5.5 to create a blockout of a village in blender, then I used FLUX 3 Klein + Fibo 2 to texture the whole map! We pick 128 views around the map and get the image model to texture gray areas, then we project the image onto the blockout. Open source here - github.com/Patch-gg/trompe
63
99
11
1,492
92,665
patch retweeted
Blendi
@BlendiBylOct 21
Prompt to 3D character animation, now with high fidelity hand + finger tracking✌️ FLUX 3 Max video → WiLoR for hands + GVHMR for body tracking → any animated character! open source: github.com/blendi-remade/pro…
Blendi
@BlendiBylOct 19
Prompt → animation at realtime speed 🕺 FLUX 3 Max generates the video, GVHMR pulls clean 3D motion from it on one GPU, and it's retargeted live onto any mixamo character. Fully automated flow! Open source: github.com/blendi-remade/pro…
9
67
4
607
26,251
with FLUX 3 Max + GVHMR, you can get any animation retargeted LIVE onto any Mixamo character! Try it out: github.com/blendi-remade/pro…
fal
@falOct 19
Introducing FLUX 3 Max, the latest post-trained video model from fal Research in partnership with @bfl_ai. FLUX 3 Max is tied for #1 in image-to-video quality on independent third-party benchmarks, generating a 10-second, 720/768p video in ~12 seconds. For the next three weeks, FLUX 3 Max is 75% off, making it the most affordable frontier video model at its quality level.
2
25
1,617
patch retweeted
Blendi
@BlendiBylOct 19
Prompt → animation at realtime speed 🕺 FLUX 3 Max generates the video, GVHMR pulls clean 3D motion from it on one GPU, and it's retargeted live onto any mixamo character. Fully automated flow! Open source: github.com/blendi-remade/pro…
fal
@falOct 19
Introducing FLUX 3 Max, the latest post-trained video model from fal Research in partnership with @bfl_ai. FLUX 3 Max is tied for #1 in image-to-video quality on independent third-party benchmarks, generating a 10-second, 720/768p video in ~12 seconds. For the next three weeks, FLUX 3 Max is 75% off, making it the most affordable frontier video model at its quality level.
17
50
7
596
50,975

AI NPCs you can actually talk to are finally here. Here's how to do them in any game

Article · @patch_gg
5
24
5
161
19,833
You can now recreate any meme in GTA V
1
6
1
126
7,343
Replying to @uratmangun2
exactly and I think with some more tweaking haiku could be at opus level for so much cheaper
1
43
Replying to @gmartidesign
none of them nailed it, they broke in opposite ways. haiku kept blender’s auto weights every run: smooth, but lift a leg and the chest moves 13cm. sonnet and opus caught the leaks and painted their own: zero bleed, but rubbery or pinched hips.
1
2
95
We compared Haiku, Sonnet and Opus 5.5 models on a simple rigging & animation task! We tracked time taken and total token cost! We'll be releasing a more in depth benchmarking and evaluation for a range of headless blender tasks!
2
4
36
2,323
patch retweeted
patch
@patch_ggOct 16
NPCs can talk back now! Introducing NPCTalk: make the NPCs of any game talk in realtime. I made every NPC in GTA V answer to me in character, in their own cloned voice, in under 1s, and they act on what you say. We packaged all the skills Claude learned into a single repo, built on open models: NVIDIA Nemotron ASR for transcription, Qwen 3.8 27B for realtime dialogue, and Fish Audio Drama 3 for the cloned voices, self-hosted on a serverless GPU. github.com/christos-antono/N…
161
597
93
9,966
838,186
patch retweeted
patch
@patch_ggOct 17
This mod is extremely fun! Talk to any NPC in GTA V
patch
@patch_ggOct 16
NPCs can talk back now! Introducing NPCTalk: make the NPCs of any game talk in realtime. I made every NPC in GTA V answer to me in character, in their own cloned voice, in under 1s, and they act on what you say. We packaged all the skills Claude learned into a single repo, built on open models: NVIDIA Nemotron ASR for transcription, Qwen 3.8 27B for realtime dialogue, and Fish Audio Drama 3 for the cloned voices, self-hosted on a serverless GPU. github.com/christos-antono/N…
12
1
90
5,329
This mod is extremely fun! Talk to any NPC in GTA V
patch
@patch_ggOct 16
NPCs can talk back now! Introducing NPCTalk: make the NPCs of any game talk in realtime. I made every NPC in GTA V answer to me in character, in their own cloned voice, in under 1s, and they act on what you say. We packaged all the skills Claude learned into a single repo, built on open models: NVIDIA Nemotron ASR for transcription, Qwen 3.8 27B for realtime dialogue, and Fish Audio Drama 3 for the cloned voices, self-hosted on a serverless GPU. github.com/christos-antono/N…
12
1
90
5,329
Load more