Flux159

11 ms·

121.
Anydoor: Zero Shot Object-level image customization(github.com)
1 points by Flux159 3y ago|1 comments
122.
by Flux159 3y ago
Hopefully Windows 12's emulation of x86-64 apps is better than the support in the Windows 11 on ARM releases - if that's the case I could definitely see the next few years being a shift for Windows laptops from x86 to ARM with Nvi
123.
by Flux159 3y ago
This would make sense since the Gallup survey is based on people's perceptions based on social & traditional media that they see regularly. People remember notable events or shocking things, no one remembers the regular days where
124.
by Flux159 3y ago
I started getting into image generation last year with Stable Diffusion's 1.5 release last October, but seeing the improvements from Midjourney and DallE 3 is extremely impressive. The image to video models from Runway and Pika labs &a
125.
Midjourney V1 to V6 – Image generation improvements from Feb 2022 to Dec 2023(twitter.com)
3 points by Flux159 3y ago|2 comments
126.
by Flux159 3y ago
The issue with this is that we don’t know when the next recession is going to happen. There could still be a recession in 2024 if the fed doesn’t get their “soft landing” while doing their planned 3 rate cuts.
127.
by Flux159 3y ago
I loved Lufia 2 as well, especially the Ancient Cave which blew my mind when I was younger since I didn't realize how they could make a full game and a replay-able 100 level dungeon in the same SNES cart. My favorite game from SNES era
128.
by Flux159 3y ago
Local AI models (LLMs, Diffusion models) will take off in popularity as their performance will be good enough for many use cases without having to use a server. Also would allow a lot of startups to leverage client computation without havin
129.
by Flux159 3y ago
I'm eagerly waiting for 7 Remake's part 2 this February, having not played the original since it was on Playstation & I only had an N64. Having played a lot of SNES JRPGs via SNES9X or ZSNES, I loved the story and character bu
130.
by Flux159 3y ago
Nice job, was going to mention url size limits but then read your blog post and it already mentioned using urls to be able to share with others without a database. One thing that might be interesting is being able to edit things in the futu
131.
by Flux159 3y ago
The numbers in the article are quite alarming for some countries - 49.9 out of 100 Canadians dependent (so one dependent for every 2 workers according to article) by 2075, or 77 out of 100 for Japan by 2075. This seems like such a critical
132.
by Flux159 3y ago
Code is at https://github.com/Kevin-thu/DiffMorpher and github page is https://kevin-thu.github.io/DiffMorpher_page/ . I think the github page shows more examples, so if it's possible to edit
133.
DiffMorpher – Using Diffusion Models for Image Morphing(huggingface.co)
1 points by Flux159 3y ago|1 comments
134.
by Flux159 3y ago
I think it’s just img2img with a prompt & rcfg scale and no controlnet since theres a GitHub issue about adding controlnet support open at the moment.
135.
by Flux159 3y ago
Arxiv paper here https://arxiv.org/abs/2312.12491 I think that it's possible to get faster than their default timings for a 4090 (I have been able to get 10fps without optimizations with SDXL Turbo and 1 iteration
136.
StreamDiffusion: A pipeline-level solution for real-time interactive generation(github.com)
365 points by Flux159 3y ago|67 comments
137.
by Flux159 3y ago
How does this compare to insanely-fast-whisper though? https://github.com/Vaibhavs10/insanely-fast-whisper I think that not using optimizations allows this to be a 1:1 comparison, but if the optimizations are not porte
138.
by Flux159 3y ago
Link to the paper directly https://arxiv.org/pdf/2310.17680.pdf The system that they describe, Codefusion is interesting because it's a diffusion model for generating code rather than an autoregressive model like
139.
by Flux159 3y ago
Online chat interface here: https://huggingface.co/spaces/HuggingFaceH4/zephyr-chat You should be able to run this locally on CPU or GPU as long as you have 16GB RAM. Less if using a quantized model like https:&#
140.
Zephyr 7B – Mistral Finetune that responds like ChatGPT(huggingface.co)
37 points by Flux159 3y ago|12 comments
141.
Show HN: Experiment with Hugging Face models in a single notebook interface(lastmileai.dev)
43 points by Flux159 3y ago|7 comments
142.
by Flux159 3y ago
ChatGPT and PaLM Chat cells do take context from previous cells of the same type (we don't mix messages between them to avoid confusion). For something like Stable Diffusion it doesn't make much sense where the prompt is going to
143.
by Flux159 3y ago
We love Jupyter notebooks, I was tech lead of an entire platform around them in a previous job - but we wanted to differentiate from Jupyter a bit by not being python first. For AI Workbooks, we thought that the interface is natural languag
144.
AI Workbooks – A notebook interface for LLMs, image and audio models(lastmileai.dev)
196 points by Flux159 3y ago|33 comments
145.
by Flux159 3y ago
Link to the github repo: https://github.com/salesforce/CodeTF It would be helpful to see some Colab notebook examples of how I could use this or incorporate my own codebase with these open source coding models. The exa
146.
by Flux159 3y ago
I think that it would be helpful to add a fine-tuning costs for an open source model (think LLaMA to Alpaca). From the phrasing around fine tuning right now it seems like it's using openai's fine tuning api to determine that cost,
147.
Using ChatGPT Plugins with LLaMA(blog.lastmileai.dev)
315 points by Flux159 4y ago|138 comments
148.
by Flux159 5y ago
I wonder what the internal data showed on these stores - from what I remember, these stores mostly had kindle devices and some selection of books which you could more easily get online. Someone probably realized that Amazon Go & Whole F
149.
by Flux159 9y ago
I like the idea as its an interesting map visualization of the subway across a time domain rather than distance. I'm wondering why there's such a large difference in certain areas between weekday rush hour and late night though (f
150.
by Flux159 9y ago
If you want to learn more about 3d engines (game or rendering), I would recommend BabylonJS: https://github.com/BabylonJS/Babylon.js It has fairly good docs: http://doc.babylonjs.com/ as well as an act

More ›