Pinned Tweet
🚨 €135 FOR A COMPLETE LOCAL AI COMPUTER. NOT JUST THE GPU. THE ENTIRE MACHINE! 🤯
And the story is even crazier.
🖥️ HP Z220, Xeon E3-1270 v2, 32GB DDR3 — €50. Bought 5 years ago. Collecting dust in my garage ever since.
🔥 AMD RX 580 2048SP, 8GB VRAM — €49 delivered this week.
💾 256GB NVMe — €30 delivered this week.
🔧 PCIe-to-NVMe adapter — €6 on Amazon.
TOTAL: €135!
We installed Linux, built llama.cpp b12318 with Vulkan, and benchmarked five quantized AI models.
📊 REAL MEASURED RESULTS
Prefill / Decode (tokens/sec):
• Qwen2.5 3B Q4_K_M — 397 / 56.8
• Nemotron 3 Nano 4B Q4_K_M — 270 / 39.1
• Nemotron 3.5 Bolt E4B Q4_K_XL — 441 / 29.3
• Ternary Bonsai 2 9B Q2_0 — 172 / 26.9
• Ternary Bonsai 2 27B Q2_0 — 48 / 9.3
Yes, even a 27B model runs on this €135 machine! Part of it is offloaded to system RAM.
Test protocol: 512 prompt tokens, 128 generated tokens, maximum GPU offload, flash attention OFF. These are benchmark workloads, not maximum context-window measurements.
GPU peak: 120W, 46°C. No thermal throttling.
🤯 AND HERE’S THE SURPRISE:
We moved the same RX 580 from a much more powerful 12-core, 64GB quad-channel DDR3 workstation into this old 4-core, 32GB dual-channel machine.
Performance barely changed!
For these small models, the GPU matters far more than the expensive workstation around it.
No cloud. No subscriptions. Your hardware. Your data.
This is a real, affordable experimental AI laboratory. And possibly a useful starting point for practical local AI applications.
🎄 Here’s an idea: build one for your kids as a New Year’s gift. Their very own computer for experimenting with language models, learning AI, and building things!
And now imagine what models we might be running on just 8GB of VRAM with llama.cpp by New Year’s Eve.
THE LOCAL AI REVOLUTION MIGHT ALREADY BE COLLECTING DUST IN YOUR GARAGE. 🚀
Looks like I’m going to make it!
Replying to @joelcomm
Honestly? There isn't one. Data shows the average founder of the fastest-growing startups is 45, and 50-year-olds often outperform 30-year-olds. Colonel Sanders franchised KFC at 62. Real limits are health, finances, and willingness to start as a beginner again—not the calendar.
36
🚀 For NVIDIA Blackwell (RTX 50xx) owners: Basalt 0.4.0 is out! The biggest update is support for its first dense model, Qwen3.8-27B. On a single RTX 5090, the developer achieved 262 tok/s in chat with speculative decoding, up to 776 tok/s on structured output, and a 262K context window.
Basalt is a fork of Strata, specifically optimized for NVIDIA Blackwell. Reworked CUDA kernels, NVFP4 and Tensor Core optimizations, faster prefill and decode. According to the developer’s benchmarks, it’s approximately 1.5× faster than Strata in text generation and 2.1× faster on long prompts using an RTX 5090.
GitHub: github.com/jesdga95/basalt/r…
Today is a day of rapid updates and bug fixes — and Basalt is no exception! 🚀
We covered Basalt v0.4.0 earlier today, released just two days ago. And now v0.4.1 is already out!
Key fixes: improved RAM management, better cache reuse in long conversations, fixed decode slowdown on some 16GB GPUs, and improved parallel request handling.
Update straight to v0.4.1!
github.com/jesdga95/basalt/r…
30
Today is a day of rapid updates and bug fixes — and Basalt is no exception! 🚀
We covered Basalt v0.4.0 earlier today, released just two days ago. And now v0.4.1 is already out!
Key fixes: improved RAM management, better cache reuse in long conversations, fixed decode slowdown on some 16GB GPUs, and improved parallel request handling.
Update straight to v0.4.1!
github.com/jesdga95/basalt/r…
🚀 Basalt v0.4.1 is now running in production on my RTX 5090 Laptop 24GB!
We benchmarked our Qwen3.8-Flash-Next IQ3_S against v0.4.0. Short-context performance is unchanged, but long-context decoding looks significantly faster and more stable.
v0.4.0 → v0.4.1 (decode tok/s):
• Short: 189.5 → 189.1
• 1.3K: 189.7 → 188.8
• 62K: 123.9 → 169.4 (+37%)
• 113K: 118.7 → 171.9 (+45%)
⚠️ Important caveat: 6 background requests overlapped with our v0.4.0 control run, so the percentage gains are preliminary, not clean A/B results.
What we can confirm: v0.4.1 was much more consistent in long-context tests (151–188 tok/s), while v0.4.0 dropped as low as 89 tok/s. Expert-cache hit rate improved from 80–84% to ~93%.
This is consistent with the --resident-budget-gib adaptive-swap regression fixed in v0.4.1 (#366/#369). Our budget is 46 GiB.
Memory essentially unchanged (~50.2 → 50.4 GiB), no swap, same temperatures. Build took just 5.5 minutes.
Update accepted. Basalt v0.4.1 stays in production!
github.com/jesdga95/basalt/r…
1
75
Already missing Strata updates? 🚀 Strata v0.2.15 is OUT — just 15 hours after v0.2.14!
This time, it’s all about the app: no more external llama-server launches, no unnecessary 5 GB draft downloads, Drafting settings now survive restarts, and smoother Qwen3.6 35B-A3B installation.
No new settings. No model changes. Just fixes.
@coldniko 🥷
github.com/Niko1221/Strata/r…
Replying to @coldniko
Keep Strata even faster and free by supporting me:
github.com/sponsors/Niko1221
buymeacoffee.com/strataengin…
Every bit counts! 🥳
1
403
Basalt 0.3.3 → 0.4.0 on P2005 (RTX 5090 Laptop), tok/s:
Short: 98→190
1.3K: 88→196
62K: 90→167
113K: 138→162
TTFT (large): 38→16s
0.3.3 previously did ~185 tok/s on short prompts, so today’s baseline seems low. Not investigating—0.4.0 is fast enough.
More to come! The open weights of smaller Qwen4 models will be released. Stay tuned. 👀
📢 Meet Qwen4-Max, your new AI teammate for work, creativity, and discovery!
Qwen4-Max is built on a new MoE architecture with native vision and a 1M-token context window.
Highlights:
- Outperforms Qwen3.8-Max with up to ~50% fewer tokens, while average cost per task drops over 60% in our internal evaluation.
- Advances real-world productivity through greater proficiency with professional tools, stronger long-horizon execution, and higher-quality deliverables.
- Handles real tasks, including 3D modeling, game development, manufacturing, finance, law, healthcare, and frontier scientific discovery.
- $1.5/M tokens for input, $4.5/M tokens for output, and $0.075/M tokens for implicit caching — all lower than Qwen3.8-Max, with implicit caching down by 70%.
The API is live on QwenCloud. You can also try Qwen4-Max on Qwen Studio, QwenWork and Qoder.
Start building today! 🚀
- Blog: qwen.ai/blog?id=qwen4-max
- API: qwencloud.com/models/qwen4-m…
- Token Plan: qwencloud.com/pricing/token-…
More to come! The open weights of smaller Qwen4 models will be released. Stay tuned. 👀
3
149
+gemini ✌️
🚨 BREAKING: GEMINI SUBSCRIPTIONS ARE COMING TO HERMES. 🔥
Today, Gemini in Hermes runs on an API key, not the Gemini plan you already pay for.
That's being worked on, the approved way:
▸ Nous is working with Google to bring your Gemini subscription into Hermes
▸ It should be officially supported, with Google's sign-off
▸ Apps plugging in a Gemini plan today aren't doing it the way Google approves
▸ Not live yet, no date
Claude and ChatGPT subscribers can already bring their plan to Hermes. Gemini is next in line.
193
🚀 Strata v0.2.14 is out! Released today, November 15, by @coldniko (Niko Veit).
This is a major architectural milestone: Strata now runs prompt-expert computation on its own native GPU kernels, replacing the previous llama.cpp path on NVIDIA and AMD GPUs (Windows/Linux).
And the benchmarks are impressive:
🔥 RTX 5090: 7,870 vs 6,851 tok/s — +14.9% prompt processing
🔥 RTX 4090: 20K-token prompt in 5.65s vs 6.11s — ~7.5% faster
🔥 RTX 5070 Ti: 1,408 vs 953 tok/s — +47.7% on a 20K-token prompt
🔥 Dual RTX A6000: 75.88 vs 64.16 tok/s — +18.3% output with improved multi-GPU memory allocation
🔥 Mac M5 Max: 94.7 vs 92.8 tok/s output with IQ3_S.
No new settings. Existing models, chats and configurations remain intact.
What makes this release particularly interesting isn’t just the speed improvement. Strata is progressively replacing inherited inference components with its own implementations.
An independent inference engine is taking shape, release by release.
Huge respect to Niko for the pace of development. We’re looking forward to testing this version on our own hardware!
👨💻 Niko:
🔗 GitHub release: github.com/Niko1221/Strata/r… teapawt.org/coldniko
Replying to @coldniko
Keep Strata even faster and free by supporting me:
github.com/sponsors/Niko1221
buymeacoffee.com/strataengin…
Every bit counts! 🥳
Bit by bit
🚀 Strata v0.2.14 is out! Released today, November 15, by @coldniko (Niko Veit).
This is a major architectural milestone: Strata now runs prompt-expert computation on its own native GPU kernels, replacing the previous llama.cpp path on NVIDIA and AMD GPUs (Windows/Linux).
And the benchmarks are impressive:
🔥 RTX 5090: 7,870 vs 6,851 tok/s — +14.9% prompt processing
🔥 RTX 4090: 20K-token prompt in 5.65s vs 6.11s — ~7.5% faster
🔥 RTX 5070 Ti: 1,408 vs 953 tok/s — +47.7% on a 20K-token prompt
🔥 Dual RTX A6000: 75.88 vs 64.16 tok/s — +18.3% output with improved multi-GPU memory allocation
🔥 Mac M5 Max: 94.7 vs 92.8 tok/s output with IQ3_S.
No new settings. Existing models, chats and configurations remain intact.
What makes this release particularly interesting isn’t just the speed improvement. Strata is progressively replacing inherited inference components with its own implementations.
An independent inference engine is taking shape, release by release.
Huge respect to Niko for the pace of development. We’re looking forward to testing this version on our own hardware!
👨💻 Niko:
🔗 GitHub release: github.com/Niko1221/Strata/r… teapawt.org/coldniko
🇪🇺 FOR SALE — European Union | Milan, Italy
I’m selling two powerful Apple computers, both purchased in spring 2026, in excellent condition and available for immediate delivery.
🖥 Mac Studio M3 Ultra
• 28-core CPU / 60-core GPU
• 256GB unified memory / 2TB SSD
• Price: €9,900
• AppleCare+ warranty and accidental damage protection until May 19, 2027, with an option for annual renewal.
💻 MacBook Pro 14” M5 Max
• 18-core CPU / 40-core GPU
• 128GB unified memory / 2TB SSD
• Price: €6,900
• AppleCare+ warranty and accidental damage protection until July 2, 2029.
✅ Both available immediately in Milan, Italy.
✅ Prices include VAT.
✅ Company invoice with VAT separately itemized.
✅ Payment by bank transfer to the company account.
✅ Prices negotiable.
Why am I selling?
My AI startup is growing, and after receiving cloud computing credits through startup programs from Amazon AWS, Google Cloud and OVHcloud, I now have more computing capacity than I currently need.
I’m selling this surplus hardware to free up funds for continuing my research, development and growth of the startup.
🙏 A HUGE FAVOR TO EVERYONE: PLEASE HELP ME SELL THESE COMPUTERS!
I really need the funds to keep working on my project and developing my startup.
Even if you’re not interested in buying, please repost and share this with anyone who might be interested.
Your support would mean a lot to me. Thank you! ❤️













