惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

MongoDB | Blog
MongoDB | Blog
罗磊的独立博客
美团技术团队
B
Blog
量子位
The Cloudflare Blog
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
aimingoo的专栏
aimingoo的专栏
The GitHub Blog
The GitHub Blog
博客园 - 聂微东
P
Proofpoint News Feed
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
D
DataBreaches.Net
博客园 - 三生石上(FineUI控件)
Y
Y Combinator Blog
酷 壳 – CoolShell
酷 壳 – CoolShell
Vercel News
Vercel News
Blog — PlanetScale
Blog — PlanetScale
云风的 BLOG
云风的 BLOG
Microsoft Azure Blog
Microsoft Azure Blog
有赞技术团队
有赞技术团队
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
阮一峰的网络日志
阮一峰的网络日志
S
SegmentFault 最新的问题

Oxen.ai

Run DeepSeek, Kimi, and GLM in Your Terminal with oh-my-pi and Oxen.ai WAN 3.0 Prompting Guide Speeding up the Oxen.ai File System MiniMax H3: Open Weights & Prompting Techniques Complete Guide. Seedance 2.5, MiniMax H3, FLUX 3, and WAN 3 are now live on Oxen! Oxen's Model Report - July 23rd, 2026 The Ultimate Guide to Topaz Upscaling Video Models: Proteus, Starlight, Wonder, and Hyperion Everything new in Oxen.ai - July 7th, 2026 Oxen.ai vs Higgsfield: pay-as-you-go AI video and image models Oxen.ai vs Magnific (FreePik): which AI creative platform should you use? Oxen's Model Report - May 8th, 2026 Writing a fine-tuning and deployment pipeline isn't as easy as it looks (Gemma 4 Version) Oxen's Model Report - April 9th, 2026 When to Fine-Tune an Image Model Oxen's Model Report - March 11th, 2026 Frank's Red Hot Bell Canada Oxen's Model Report How a $1 Qwen3-VL Fine-Tune Beat Gemini 3 How to Train a LTX-2 Character LoRA with Oxen.ai How to Use WAN 2.1-VACE to Generate Hollywood-Level Video Edits How We Cut Inference Costs from $46K to $6.5K Fine-Tuning Qwen-Image-Edit How to Set Noise Timesteps When Fine-Tuning Diffusion Models for Image Generation Fine-Tuned Qwen-Image-Edit vs Nano-Banana and FLUX Kontext Dev
Isometric.nyc
Eloy Martinez · 2026-02-21 · via Oxen.ai

A giant isometric pixel-art map of New York City, inspired by SimCity 2000 and
Rollercoaster Tycoon. Andy Coenen fine-tuned an image model on Oxen with just 40 training examples to nail the retro aesthetic, then built a pipeline that
generated the entire city tile by tile, over 200 generations per hour at under
$3/hour. The full project and writeup lives here.

We love seeing creators push fine-tuning in unexpected directions. If you're an
artist or brand working on AI-driven projects and want help with fine-tuning,
reach out at hello@oxen.ai