Quan Tran

Fears small losses. Misses big wins.

[Quan Tran @Fech_oScratch]

So I made this film using Unreal!
The Epic Games Unreal Fellowship Winter 2021 Program recently wrapped.
I was privileged to work alongside some amazingly talented film makers from around the world.
https://t.co/vUvMbwV6dd

[Quan Tran @Fech_oScratch]

BTC holding $86k while 38/100 CD100 names bleed red = classic thinning breadth. Meanwhile BCH ripping 32% on that CME futures listing shows liquidity chases catalysts, not conviction. Brent under $100 adds a macro tailwind. Rotation or exhaustion? #Bitcoin

[Quan Tran @Fech_oScratch]

BTC chilling at $86k while BCH pumps 32% on a CME listing? Wild how liquidity keeps rotating into the most random corners of this market. Is anyone else surprised BCH is the one stealing the show today? #Bitcoin

[toyxyz @toyxyz3]

I made a new custom node for more precise control over camera and character movements in MiniMax-H3! https://t.co/XVs5CO3Q46

[Quan Tran @Fech_oScratch]

@fyreinteractive Course

[Quan Tran @Fech_oScratch]

@natecurtiss_yt Script

[Jacob Coxon @hilbertspaess]

I resigned from Anthropic today. I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives. More thoughts below.

[silvan @silvanrec]

SIGAN A ESTA PERSONA SI QUIEREN HACER DINERO, LO DIGO ENSERIO.

[MAX @maxxmalist]

you can one-shot your ads even faster now with GPT-6 Astra

here’s the full workflow for realistic UGC:

- give Astra your product details, audience, offer and example ads
- give it browser access and tell it:

“open meta ad library, search these 5 competitors in the us and shortlist 10 active ugc ads, prioritize longer-running ads and recurring angles, save the links and break down each hook, pain point, product demo, objection and cta, don’t assume runtime proves profitability”

- ask it to turn the research into 3 original concepts for your product, using the patterns it found without copying the scripts
- have it write the strongest concept as a casual conversation, with a clear hook, problem, product demo and cta, then split it into short scenes with exact dialogue and actions
- generate first frame images for your creator, product and setting, then reuse them across scenes to keep the face, outfit, packaging and room consistent (add iphone 17 pro quality to the prompt to maintain realism)
- connect Astra to your generation tools through a CLI, API or just browser access
- ask astra to write prompts for realistic phone footage, with natural skin texture, imperfect framing, subtle handheld movement, small gestures and conversational pauses
- have it assemble the clips and add captions, then review the full ad for lip sync, voice changes, weird hands and distorted product details
- save the research, references, prompts and workflow in markdown files, then give astra that folder with your next brief

once your MD file will be ready you can just upload it to new chat and Astra will immediately know what to do step by step (or simply copy paste this tweet along with your manual workflow in bulletpoints)

set it up once and your next request can be ��make 3 new ugc ads for this product using our saved workflow”

bookmark this for later, more details are in the article

[Hunter Weiss @Hunter_Weiss]

I spent all day editing with Davinci MCP in Claude. Wow it’s insane. Testing Astra tomorrow. I probably saved 10 hours today. Mind blowing stuff from someone who’s edited for 15+ years across various software. It’s only going to get better. It did such crazy color grading on raw Alexa footage I was mind blown

[Quan Tran @Fech_oScratch]

@tye1337 Trust

[gokaygokay @gokayfem]

GPT-6 Astra and H3 Max combo is insane!

I have made an insanely fast renderer inside blender

Astra is drawing the geometry in realtime and H3 Max is rendering the updates extremely fast

[tapehead.lab @tapehead_Lab]

MiniMAX H3で出したよくわからないものを、Fable5.1でさらによくわからなくした

[Quan Tran @Fech_oScratch]

@Word213414 Outdated

[Quan Tran @Fech_oScratch]

@julianfaceless Launch

[siddharth ahuja @sidahuj]

Take a photo of the real world
Paste it into Blender 3D

[Jonny Miller @jonnym1ller]

a brief love letter to codex https://t.co/Tq4VEUHoIu

[Annatar.md @AnnatarXBT]

NVIDIA is giving away free access to 140+ AI models for a full year

> a lot of people building agents are paying $50-200/month just for API access

NVIDIA just made that argument irrelevant

models you get: GLM 5.2, MiniMax M3, Nemotron-3.5-lightning-30b-a3b, Muse-glimmer-30b from Meta and 130+ more

setup:

> step 1 - get your free key
> go to https://t.co/xscD1lTuhg
> register -> bind phone -> copy API key

> step 2 - add to Hermes agent
> open Settings -> Model Provider -> Custom

base_url = "https://t.co/R7kYRIgm4r"
api_key = "nvapi-xxxxxxxxxxxxxxxxxxxx"

> step 3 - pick a model

model = "z-ai/glm-5.2"
model = "minimaxai/minimax-m3"
model = "nvidia/nemotron-3-ultra-550b-a55b"
model = "moonshot-ai/kimi-k2.7"

> Hermes already has NVIDIA set as default base_url
> paste the key and you're running instantly
> works the same in Cursor and OpenCode

> cost: $0
> limit: 40 req/min
> expires: 1 year

a free key changes what you can afford to run:

> one prompt costs nothing either way
> a loop firing 300 agents every week is where the bill shows up
> for a year that bill is zero

while everyone is paying for API access, this is sitting there for free with this many models in one place

it is also the easiest way to test which one actually fits your agent before you commit to a paid provider

[Pietro Schirano @skirano]

I had early access to GPT-6. This model will completely shatter your understanding of what's possible, and it will start a new era of creativity.

It's hard to explain how big of a jump this is, so I'll share my tests.

1/6 Made this 3D model and animation in code from an image.

[Cristóbal Valenzuela @c_valenzuelab]

We are sharing a major update to our General World Models efforts. GWM Worlds 2 can generate full, interactive, real-time video simulations in one continuous 720p stream at 24 fps and audio at 48000 Hz, responding to your inputs as you explore.

It is the first world model that generalizes to arbitrary actions rather than a fixed set of actions or just navigation. Dynamic actions are what actually matter for building gaming and interactive experiences, and for training agents in these simulations.

We also introduced WorldPrompt, a way to tell GWM-2 that a world can be split into two kinds of state: what persists and what changes over time. For example: persistent gravity, physics, and light, alongside actions that describe movement, gestures, object interactions, speech, and sound.

[David Pantera @davidpantera_]

With Atlas, you can literally make films.

And it also basically solves first-frame-last-frame transitions.

Reconstruct a 3D scene from images, then position the camera exactly where you want it, and let Atlas fill in what the camera would see along that path.

You’re in the director's chair, staging the scene.

Here we made a 1 min video with a couple images of totally different scenes, and we set where the camera keyframes should move through each part of the scene. The scene-to-scene transitions are great.

Atlas makes it feel like a smooth shot with no cuts.