NullFoundry (@nullfoundry) profile picture

software maker. building weird things with AI, local LLMs and backend systems. shipping, breaking, benchmarking, repeat.

somewhere
Joined October 2019
Photo by NullFoundry (@nullfoundry)Photo by NullFoundry (@nullfoundry)Photo by NullFoundry (@nullfoundry)Photo by NullFoundry (@nullfoundry)Photo by NullFoundry (@nullfoundry)Photo by NullFoundry (@nullfoundry)Photo by NullFoundry (@nullfoundry)Photo by NullFoundry (@nullfoundry)Photo by NullFoundry (@nullfoundry)
NullFoundry retweeted
Strata is really the best surprise for local open source models. It makes Qwen Flash Next very usable in consumer hardware. In a 5090 it's almost comfortable.
Strata Strata 0.2.6 is out with a nice performance boost for local DeepSeek. On a single R9700 with 251 GB RAM, decode speed went from 24.5 to 27.6 tok/s in the release tests. Loading takes longer, but inference gets faster. Not a bad tradeoff. github.com/Niko1221/Strata
3
3
70
3,437
NullFoundry retweeted
🚨🚨ALERT: ANTHROPIC IS A/B TESTING USAGE LIMITS Many users on X are reporting LOWER usage limits right now. THEY ARE NOT WRONG. We detected an A/B test going on that DECREASES your monthly API-value by ~20% while INCREASING your 5h usage limits by ~20%. This only affects SOME accounts. (1/2)🧵
201
199
163
3,234
523,278
NullFoundry retweeted
Those two 🤣
195
714
183
7,741
758,356
NullFoundry retweeted
Do you hear this ! Qwen is working on a 35B checkpoint !!! This is what all of us with low vram, 3050, 3060, 3080 are dreaming for! There is a good chance we will get Qwen3.8/4.x-35B-A3B now 🤩 Let's make some noise, let @QwenDevs hear how much we appreciate their work, and more importantly, how much we are looking forward to new 35B 😍 Special thanks to @Biggest for communicating what we local ai folks love to see, with them.
I'm doing everything I can to get this released. They're actively testing and training an A3B MoE model checkpoint. glad you and others are keeping up the pressure publicly. genuinely helps! i'm hopeful.
33
27
6
681
38,581
Day 175 making my own TES until TES 6 comes out. Testing_Bow_&_Explosion #elderscrolls #indiegamedev #pixelart #daggerfall
41
70
4
1,095
45,612
NullFoundry retweeted
AI-MEMORY is the premiere Long-term memory for AI coding agents. Quit Claude Code mid-task, start Codex in the same directory, continue without re-explaining the architecture, the failed approaches, or the open questions. With proper memory consolidation, Wiki-based documentation accessible to your team. Multi-user support for shared memory and much more. It support all main harnesses, all operating systems. Run stand-alone in your machine for single-user, run self-hosted for your team, all open source, no shady cloud-backing necessary: you own your knowledge. This project is constantly monitoring all main competitors and adjusting and improving to keep being the best option. Just updated a new section in the README to summarize key points: github.com/akitaonrails/ai-m…
32
32
4
381
59,984
NullFoundry retweeted
I interviewed DHH, the creator of Omarchy - The Agent Operating system + Ruby on Rails, one of the most important frameworks of our time. He's a father, builder, race-car driver, and such an inspiring person. One of the most positive people I've had the honor of meeting.
89
135
34
1,816
203,138
If AI mapped a cure for terminal pancreatic cancer I don't think anyone would dare complain about its effects on medical unemployment. But people are upset that AI can solve math problems and cause unemployment in performative mathematics. Fuck them.
74
828
128
8,751
395,228
NullFoundry retweeted
This game is entirely "vibe" coded. I spent 6 months using AI to replace Unreal Engine from under our game Em-8ER. Now we're developing 93x faster than we ever did before with Unreal. Because our engine is developed to be 100% AI first instead of tacked on. It is not a toy one shotted overnight, it is a production ready engine. I made a big bet that this was the future and now I'm very confident it is. You don't need to license an expensive game engine anymore. Today's update features: - Advanced character animation system - World collision meshes - AI assisted Blender anim exporter and workflow - Character outfits and outfit swapping - Physic driven character effects - Fully networked client/server player capsules - Pilotable vehicles - Waifus The engine is called Kun-Zhi, the Tibetan word for "Ground of all" that the world springs from. This engine will be released to the public, starting with backers first. Today we're opening up to public backers again after a quiet building period where they were closed. If you would like to support our engine work and get the game, you can subscribe as a patron, or purchase a game backer pack on sale at 50% here: my.em8er.com/perks Corporate sponsorships are available. Send a DM. Current partners include Razer and AMD. Follow and subscribe for our first "How To" articles on AI coding and gaming dropping tomorrow.
1,336
578
423
10,537
3,220,862
NullFoundry retweeted
🫶Let’s keep evolving
The people's AGI. Thank you Qwen.
74
48
6
1,901
105,701
NullFoundry retweeted
Welcome to the Future!
10
26
7
328
32,120
NullFoundry retweeted
📢Meet Qwen3.8-Max — our most capable model to date. Next week, the open weights of Qwen3.8-Max will be released, and Qwen3.8-27B is also going open-weights to meet you all!🎉 Qwen3.8-Max, a new bar for coding and cowork at 2.4T parameters: - Autonomous coding: 10+ days of self-evolving development, from empty folder to production without hand-holding, complete project trace in the GitHub:github.com/qwen-code-dev-bot… - Real work, real results: Production-quality deliverables across hundreds of professions. - Long-horizon mastery: System-level autonomous planning with closed-loop adaptive learning, driving 500+ turns of chip design optimization and 365 days of e-commerce strategy. - Native multimodal intelligence: Vision isn't just input — it's a continuous feedback loop for planning, execution, and self-correction. 💰Pricing: Input: $2.0 / M tokens Output: $6.0 / M tokens Implicit Caching: $0.25 / M tokens Start building with Qwen3.8-Max! 🚀 📖 Blog: qwen.ai/blog?id=qwen3.8 ✅ Qwen Studio: chat.qwen.ai/?models=qwen3.8… ⚡ API: qwencloud.com/models/qwen3.8…
1,225
2,773
1,982
22,506
7,268,502
NullFoundry retweeted
Qwen3.8 is launching and going open-weight soon!🌐 With a massive 2.4T parameters, this model is continuously evolving. We believe it’s one of the most powerful model available today, compatible to leading frontier AI models , second only to Fable 5. You don't have to wait to test it. Just now, the Qwen3.8-Max-Preview made its debut on Alibaba’s Token Plan, Qoder, and QoderWork. Be among the very first to try it out. Can't wait to hear what you build. Stay tuned! 🚀  Token Plan international:qwencloud.com/pricing/token-… China:platform.qianwenai.com/prici…
1,305
3,249
2,058
24,014
8,437,218
NullFoundry retweeted
Kimi K3 subway FPS result, some say it's benchmaxxed due to my influencerstatus™️ idk man, idk...
32
42
2
798
49,650
NullFoundry retweeted
“TRACE: Capability-Targeted Agentic Training” got Spotlight @ ICML AIWILD 🎉 Beats direct RL, GEPA, & synthetic-agent data on SWE-Bench Verified and τ²-Bench. TRACE-Qwen3.6-27B tops GPT-5.2-Codex, GLM 5, & Claude 4.5 Sonnet on SWE-Bench. Co-led with @TarunSures41845. Thanks to @JonSaadFalcon and our advisor @Azaliamirh. Details below 👇
13
35
9
242
174,528
NullFoundry retweeted
🖥️Damn Local AI community. Google comes through. Gemma 4 is now fully on-device in React Native! 🎯 Run Google’s Gemma 4 100% offline in your cross-platform mobile apps with hardware acceleration: ⚡ Vulkan delegate on Android ⚡ MLX delegate on Apple Silicon 🏃‍♂️How to get started: ✦︎ Use the open-source library: react-native-executorch ✦︎ Load models easily with the useLLM hook ✦︎ Supports vision + tool calling locally 💰Cost: ✦︎ $0 API cost — fully on-device ✦︎ Main cost = device performance (best on newer phones) Best for: ✦︎ Vibe-coded React Native apps that need strong local AI with privacy and no internet dependency. 🔗 Demo & Docs: github.com/software-mansion/…
Gemma 4 now works on-device using React Native! You can now run Gemma 4 fully offline in your cross-platform apps with local hardware acceleration: ⚡ Vulkan delegate on Android ⚡ MLX delegate on Apple Silicon See Gemma 4's vision and tool-use capabilities in action, instantly reading a flyer and scheduling a calendar event, 100% on-device.
4
8
52
7,385
NullFoundry retweeted
🚨 BREAKING: Alibaba just dropped a vector database that could change RAG forever. Meet Zvec ✅ No server. No Docker. No cloud bills. ✅ pip install zvec and you're ready in under a minute. ✅ Dense + sparse + hybrid search in one API. ✅ Searches billions of vectors in milliseconds. ✅ 100% Open Source (Apache 2.0). If you're building AI apps, this is worth checking out. GIVEAWAY: I'm giving 10 people access to my favorite AI resources & tools. To enter: Like & Repost + Comment "ZVEC" Winners will be announced in 48 hours. 🔗 Link in the first comment.
90
229
12
1,642
133,271