frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

We trained a real-time world model for $2k with Minecraft mod revenue

4•dnlayux•1h ago
Hey HN, this is Daniel and Artur from DreamForge. We’re two friends and students from Kazakhstan who trained a real-time controllable world model (first-person + third-person) end-to-end for roughly $2k by standing on open models and limited revenue from our game mods. You can see early demos here: https://trydreamforge.com.

The slightly strange part is our small budget and how it was even formed. We come from several years of indie gamedev and game modding background. We’ve got mods with millions of downloads that have been covering a meaningful chunk of our cloud GPU costs. So, the budget constraint was the thing we’ve been building around this whole time.

Our first model (DF-World 0.1 Preview) is 1.5B parameters in size, generates video at 480p, supports real-time mouse+keyboard control in both first-person and third-person view, accepts mixed multimodal inputs (text, image, video), and lets you reprompt mid-stream to change the world on the fly. It runs in ~10-12 FPS on a single RTX 4090 in fp8 (~4GB VRAM), and even works on my RTX 2060 laptop, fully locally. DF-World 0.1 allows for infinite rollouts, though error accumulation changes the world too much after a minute of streaming.

As we couldn’t afford to train big models from scratch, we took LongLive 1 (autoregressive finetune of Wan 2.1 1.3B DiT) as the video backbone, then adopted the residual action module architecture from Matrix-Game 2 and trained it to work with dual-perspective control. We built a custom causal runtime layer on top of LongLive pipeline, added zero-shot multimodal conditioning, and also optimized heavily for low VRAM and consumer devices.

We still made some big strategic mistakes, for example, spent too long perfecting the video generation capabilities before video control was even established, and kept iterating/refining instead of shipping much earlier. Thankfully, all the great open-source releases kept bailing us out. The result is still clearly behind world models from frontier US/Chinese labs, but it’s a working real-time controllable loop, the first step in our world modelling attempt. For us, this feels like the natural next step from modding and gamedev, just instead of hand-building every world we made the model that generates them in real time.

Nevertheless, current limitations are obvious: DF-World 0.1 has weak spatial memory, error accumulation after ~1 minute, and latency that’s realtime but not comfortable/playable yet. But we are not claiming a breakthrough in the area of world models, so this is just the first visible loop of something we want to keep pushing further.

Right now, we are already building the next DF-World 0.5 model, and plan to open-source it once it’s less rough. We’d really love to hear some honest feedback on the current pre-recorded demos on our website (we couldn’t provide on-demand demos due to budget limitations), and your overall thoughts on DreamForge.

Rust Commercial Network Launches to Bring Commercial Rust Users Together

https://rustfoundation.org/media/rust-commercial-network-launches-to-bring-commercial-users-of-ru...
1•thesuperbigfrog•1m ago•0 comments

Ebola confirmed in France: Officials report first case of eye-bleeding disease

https://www.dailymail.com/health/article-15925635/Ebola-virus-confirmed-France-doctor-tests-posit...
1•Bender•1m ago•0 comments

LLM Research Is Fake

https://www.quoin.ai/
1•quoinai•1m ago•0 comments

Security Is Inherently Political

https://blog.miloslavhomer.cz/security-is-a-political-problem/
1•ArcHound•3m ago•0 comments

Flock CEO: "They're Gonna Enforce Immigration No Matter What Flock Does"

https://ipvm.com/reports/flock-ceo-immigration
2•jhonovich•7m ago•1 comments

Ubisoft co-founder Claude Guillemot dies in plane crash

https://techcrunch.com/2026/06/21/ubisoft-co-founder-claude-guillemot-dies-in-plane-crash/
5•randycupertino•7m ago•0 comments

Where might we find life in our solar system?

https://www.nationalgeographic.com/science/article/where-might-we-find-life-in-our-solar-system
2•bookofjoe•8m ago•1 comments

Cerebras CEO says margin forecast misunderstood as stock plummets after earnings

https://www.cnbc.com/2026/06/24/cerebras-cbrs-stock-earnings.html
4•mikhael•9m ago•0 comments

Microsoft's quantum computing technology called into question, again

https://www.reuters.com/legal/government/microsofts-quantum-computing-technology-called-into-ques...
2•tartoran•9m ago•0 comments

Trump cancels bipartisan housing bill, reiterates demand for SAVE America Act

https://www.cbsnews.com/news/trump-signs-housing-bill-capitol/
5•mikhael•10m ago•0 comments

Raspberry Pi and the EU Cyber Resilience Act

https://www.raspberrypi.com/news/raspberry-pi-and-the-eu-cyber-resilience-act/
2•Brajeshwar•10m ago•0 comments

Stripe, Anthropic, and OpenAI are backing an effort to stop respiratory infecti

https://www.technologyreview.com/2026/06/24/1139621/stripe-anthropic-and-openai-are-backing-an-ef...
3•joozio•10m ago•0 comments

You Don't Know Jack About Formal Verification

https://queue.acm.org/detail.cfm?id=3819084
2•yarapavan•11m ago•0 comments

LLM from Scratch: a small LLM running inside MIT's Scratch

https://github.com/Broyojo/llm_from_scratch
3•alexkranias•11m ago•0 comments

DOD: Grok was used to fire thousands of missiles in the Iran war

https://www.utilitydive.com/news/doj-intervenes-xai-data-center-gas-turbine-lawsuit/823267/
2•skolos•13m ago•1 comments

OpenClaw-memgpt – MemGPT's three-tier memory architecture as an OpenClaw plugin

https://github.com/xltvy/openclaw-memgpt
4•xltxy•13m ago•0 comments

Web-Based Indirect Prompt Injection Observed in the Wild

https://unit42.paloaltonetworks.com/ai-agent-prompt-injection/
2•wslh•14m ago•0 comments

The Promptware Kill Chain

https://arxiv.org/abs/2601.09625
2•wslh•15m ago•0 comments

Microspeak elaborated: Isn't escrow just a release candidate by another name?

https://devblogs.microsoft.com/oldnewthing/20260623-00/?p=112462
5•AndrewDucker•17m ago•0 comments

The Three AImigos versus the Magnificent Seven

https://www.ft.com/content/98582580-20ed-4205-9ef2-4d56657b0004
2•aanet•17m ago•1 comments

Nothing has killed my GTA 6 hype faster than locking a core part of its identity

https://www.pcgamer.com/games/action/nothing-has-killed-my-gta-6-hype-faster-than-locking-a-core-...
2•evo_9•18m ago•0 comments

Don't Put That PDF in Your Event

https://docs.eventsourcingdb.io/blog/2026/06/25/dont-put-that-pdf-in-your-event/
2•goloroden•20m ago•0 comments

LLMs and Performative Productivity

https://joshcollinsworth.com/blog/productivity
4•birdculture•22m ago•0 comments

Show HN: Catbath: Terminal-first, browser-curious, and extension-friendly editor

http://sekor.eu.org/catbath/
2•modinfo•22m ago•0 comments

Malicious AI 'Skills' on OpenClaw's ClawHub Marketplace Bypass Scanners

https://cyber.netsecops.io/articles/openclaws-skill-marketplace-and-the-emerging-ai-supply-chain-...
5•jaybode•22m ago•0 comments

Nobody Here Wants the Data Center: An Oral History

https://newrepublic.com/article/211363/ai-data-center-oral-history
4•speckx•23m ago•0 comments

German federal Microsoft spending hit €481M in 2025

https://www.heise.de/news/Bund-in-der-Abhaengigkeitsfalle-Kostenexplosion-bei-Microsoft-Lizenzen-...
2•logickkk1•24m ago•0 comments

A green unit test, a red CI, and a chown that raced a zsh lock file

https://inferhaven.com/blog/2026-06-18-building-haven-bench-in-the-open/
2•inferhaven•25m ago•0 comments

Colorado River's Dire Water Picture Expected to Hit Green River Ranchers Hard

https://cowboystatedaily.com/2026/06/23/colorado-rivers-dire-water-picture-expected-to-hit-green-...
3•Bender•25m ago•0 comments

Luck Surface Area

https://blog.danwald.me/luck-surface-area
3•danwald•27m ago•0 comments