BLACKFURY Exact V1.0.1 Released - built on #TensorFold by @ashxhart 🚀
Qwen3.8-27B NVFP4 on one RTX 5090: 309.53 tok/s decode at 131K context in a single stream
+10.20% vs the previous release, comparing medians from separate runs.
Code 👇
Qwen3.8-27B NVFP4 + DFlash2 running locally on a single RTX 5090 - powered by #TensorFold 🔥
⚡ 753.3 tok/s peak live
📊 376.5 tok/s average decode across 36 requests 🧠 Thinking LOW
🛠️ Real coding, browser checks and visual fixes
Peak at 5:21 - even Goku powers up 😂
71
734 tok/s coding peak on ONE RTX 5090 🔥
330 tok/s average decode in the same loop.
BLACKFURY Exact: my #TensorFold variant optimized for single stream.
Qwen3.8-27B NVFP4 + DFlash2. 128K window.
Peak sampled over ~1s.
Built on @ashxhart's work. Code + results below 🧵
RTX 5090. 12 branches. Over 1,000 tok/s🤯
Qwen3.8-27B NVFP4 locally on my experimental TensorFold/OMP.
🚀 1,081 tok/s peak aggregate
⚡ 791 tok/s average across the generation batch
🧠 Thinking ON, LOW effort
💾 64K-context
Local AI is getting wild @ashxhart and #TensorFold 🔥
One RTX 5090. Qwen3.8-27B NVFP4. Eight synchronized branches sharing one context.
With thinking set to low, I measured 499 tokens/s aggregate, including reasoning, with a 716.3 tokens/s peak over a roughly one-second window, with 66k context
In this run, the agent generated eight Three.js components, integrated them and passed the browser checks in 101 seconds.
Running locally with #TensorFold + #OMP. Live coding and the result below 👇
1
2
109
Replying to @ashxhart
@ashxhart I've been playing with #Tensorfold and i wanted to replicate your experiment. Mate I'm blown away. Peak on Qwen 27b NVFP4 900+ TK/s and 661 aggregate. I feel like i can push it more.
We are 18% from the 24k goal. Let's go !!!
Any support is welcome. Go to his post and distribute it.
You can support Ash here ko-fi.com/ashxhart/goal?g=12
Following my earlier post about the M5 Ultra 512GB Mac Studio, more than two dozen people asked me to set up crowdfunding for it.
I always try to pay my own way, so asking for help at this scale doesn't come naturally. But I couldn't make the numbers work this time, and my loan got rejected after placing the order.
I'd love to get one for TensorFold and to see how far we can push it.
If you'd like to help towards it, I've put the goal on Ko-fi below. Please only contribute if you're comfortable doing so.
ko-fi.com/ashxhart/goal?g=2
Thank you to everyone who's encouraged me to do this, and to everyone already supporting TensorFold. I'm incredibly grateful and will continue to push TF. ❤️
1
4
319
160 GB of Corsair Vengeance DDR5 RAM, made up of two kits:
• 96 GB — 2×48 GB, rated at 6000 MT/s, CL36
Model: CMK96GX5M2E6000Z36
Revision: 4.53.02
• 64 GB — 2×32 GB, rated at 6000 MT/s, CL30
Model: CMK64GX5M2B6000Z30A2
Revision: 5.43.01
All four modules currently run together at 3600 MT/s, for a total of 160 GB.
1
29
Replying to @QwenDevs @KyleHessling1
I got similar speeds actually sustained across the full context
1
185
Pull it. Let go. Hit record.
Elastic glass, spring motion, and light that follows the sound.
A closer look at Liquid Glass Recorder.
2
77
Boldt Castle Power House update 🏰
Astra is killing it 💪🏻
Reworked the roofs, clock tower and stonework using reference photos. Clearer water, improved foam and backwash.
Built in three.js. Six cinematic angles.
What should refine next?
#threejs #CreativeCoding
1
77









