mathi0750

Optimizing Qwen-Image-Edit to Generate 1.2M Images

Posted by mathi0750 on Wednesday October 22, 2025 @03:27PM
from the self-post dept.

We're building off past Qwen-Image-Edit Fine-Tune Fridays and pushing the limits for 1.2 million images. We'll showcase a real world example where we cut an inference budget for a customer by nearly $50k on just one of their runs using Baseten for compute. Here's the luma if you want to join the live stream: https://luma.com/fine-tuning-friday-10

( Read More... | no comments | 2 points | on HN )

Re:The Future of Large Files in Git Is Git on Friday August 15, 2025 @09:24PM (#44919197)

Have you tried Oxen.ai? they are doing more fine-tuning and inference now but they have an open-source data version control platform written in rust at the core of their product.

Re: on Friday June 27, 2025 @10:33AM (#44397116)

Hey folks, With FLUX.1 Kontext [dev] dropping yesterday, we're comparing prompting it vs a fine-tuned FLUX.1 [dev] and PixArt on generating consistent characters. Besides the comparison, we'll do a deep dive into how Flux works and how to fine-tune it. What we'll go over: * Which models performs best on custom character gen. * Flux's architecture (which is not specified in the Flux paper) * Generating synthetic data for fine-tuning examples (how many examples you'll need as well) * Evaluating the model before and after the fine-tuning * Relevant papers and models that have influenced Flux * How to set up LoRA effectively This is part of a new series called Fine-Tune Fridays where we show you how to fine-tune open-source small models and compare them to other fine-tuned models or SOTA foundation models. Hope you can join us later today at 10 AM PST! https://lu.ma/fine-tuning-friday-3

Re:Live Dive into How to Finetune DeepSeek R1 on Synthetic Data on Wednesday January 29, 2025 @10:20PM (#42874453)

have you tried it on perplexity yet?

Live Dive into How to Finetune DeepSeek R1 on Synthetic Data

Posted by mathi0750 on Wednesday January 29, 2025 @10:15PM
from the oxen.ai dept.

https://www.oxen.ai/entry/fine-tune-deepseek-r1-with-synthetic-data

( Read More... | 2 comments | 3 points | on HN )

Show HN: Beta Model Eval Tool – Run SOTA Models on Your CSV

Posted by mathi0750 on Thursday January 16, 2025 @05:42PM
from the show-hn dept.

https://www.oxen.ai/entry/model-evaluations-reimagined

( Read More... | no comments | 5 points | on HN )

Re:Show HN: GitLFS taking forever? No prob, heres the best AI data ver... on Tuesday December 31, 2024 @02:39PM (#42561202)

Thank you:)

Show HN: GitLFS taking forever? No prob, heres the best AI data versioning tools

Posted by mathi0750 on Monday December 30, 2024 @04:52PM
from the show-hn dept.

https://ghost.oxen.ai/the-best-ai-data-version-control-tools/

Hi HN!I just released a blog on the fastest, easiest, and most collaborative AI data version control tools out there.If you’ve been iterating on AI datasets lately, you know how annoying and difficult it is to version and share your data. If you want a detailed comparison on the best data exploration and versioning options from Oxen.AI (where I am currently building a solution), Hugging Face,...

( Read More... | 2 comments | 5 points | on HN )

Re:Ask HN: How do you version your data? on Thursday December 26, 2024 @08:01PM (#42519130)

I just released (1 minute ago) a blog going into the most popular data version control options and comparing them. Hopefully this clarify which is the best solution for you. Heres the blog - https://www.oxen.ai/blog/the-best-solution-for-ai-data-versi...

Re:We put 1M files into DVC, Git-LFS, and Oxen.ai on Tuesday November 05, 2024 @02:21PM (#42054406)

Oxen is awesome. Been having a lot of fun with your model inference tool. any time line on new models + which models will you guys be adding next?