niklaslenz.com
niklaslenz.com is Play the world in your browser. Every number, the exact command and the raw output. It is ranked #13 on the AI Radar, in Open & local models, first seen 18 days ago and shared in 104 posts (345.3k views).
What niklaslenz.com says about itself
Mia's TensorFold recipe: Qwen3.8-Flash-Next, NVFP4 + MTP on 128 GB: Speed v2, Quality v2 and build test, every number with raw output and exact command.
Mia's TensorFold recipe: Qwen3.8-Flash-Next, NVFP4 + MTP — Results v2 — niklaslenz.com niklas lenz Compare Tools Method @niklaslenz_ai Home / 128 GB 128 GB · 1× DGX Spark (GB10) Mia's TensorFold recipe: Qwen3.8-Flash-Next, NVFP4 + MTP 1x NVIDIA DGX Spark Recipe (one GB10 box) Mia's AI Lab X @MiaAI_lab GitHub Hugging Face Website TensorFold (Zig, tensorfold-native) image tensorfold-qwen38:zig-db28187-c0a2df4c680d (built locally from recipe @656879fd, TF_REF db281878, patches c0a2df4c680d), MTP drafts on (depth 6) commit recipe 656879fd 1 GPU measured 2026-10-20 independent check passed Compare with another run Rank in 128 GB 1 of 9 Overall…
What people said about niklaslenz.com on X
This is so cool! Idol turns one image into a 3D model in 14.5 s on my RTX 5090. A 16 GB RTX 5060 Ti does it in 57.4 s. RTX 5090 (32 GB): Per image, 262,144 Gaussians: 14.5 s Per image, 1,048,576 Gaussians: 15.1 s First image, compile included: 22.1 s Peak VRAM: 7.5 GiB RTX 5060 Ti (16 GB): Per image, 262,144…
— @niklaslenz_ai, 5 days ago · 708 likes · see the post
🚨 DeepSeek-V4.1-Flash-1017 on one RTX 5090 🚨 That's a 748B model on a single card. His Maya-S quant, 340 GB on disk: the card holds 1,970 of its 15,360 experts, the rest is read from 96 GB of RAM and the NVMe SSD. One slot. Agent: DeepSeek Harness 0.2.1-rc.2. One prompt, one try, no help. Thinking on, sampling from…
— @niklaslenz_ai, 13 days ago · 205 likes · see the post
Absolutely mind-blowing result from DeepSeek V4.1 Flash 1017! 🔥 This is the power of full official weights! Make no mistake, every quantization degrades quality, even if it's small! IF you can use the official weights, no reason NOT TO! Get it here:…
— @MiaAI_lab, 10 days ago · 226 likes · see the post
Alternatives to niklaslenz.com
- kimi-k3.1-pro — Releasing the model weights of Kimi K3.1 Pro
- unsloth.ai — Guide to running Mistral Large 4 GGUFs locally with Unsloth and llama.cpp.
- project-maya — Plus a round of fixes, UI improvements and community PRs.
- twil-8b — Text Generation model with 9B parameters. Download, explore files and run it on Hugging Face.
- deepseek-harness — I have all three of my AI agent leads for Hermes, Grok Bot, and OpenClaw pouring over the repos for DeepSeeks proposed…
- deepseek-v4.1-flash-1017-gguf — Image-Text-to-Text model with 748B parameters. Download, explore files and run it on Hugging Face.
niklaslenz.com in numbers
- Rank on the AI Radar: #13 of 10384
- Shared in 104 posts by 8 accounts: @niklaslenz_ai, @MiaAI_lab, @ashxhart, @plotarmordev
- 345.3k views on those posts
- First seen 18 days ago, last shared 7 h ago
- Pricing seen by AGK Intelligence: not stated
- Market: Open & local models
FAQ
What is niklaslenz.com?
Play the world in your browser. Every number, the exact command and the raw output It was first shared on X 18 days ago and is ranked #13 on the AI Radar.
Is niklaslenz.com free?
Pricing is not stated on the page we read.
Who shared niklaslenz.com?
8 accounts on X, including @niklaslenz_ai, @MiaAI_lab, @ashxhart, in 104 posts totalling 345.3k views.
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 65.7k posts from 5k X accounts over the last 14 days, 10.4k tools, 19 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-06 12:45 UTC. Full method.