Top all-time · #5041
Brev.dev Console
Company blog · brev.dev/blog · on HN as naderkhalilaiprogramming
Brev.dev Console is the company blog at brev.dev, writing about ai and programming. Since 2022, 15 different Hacker News readers have linked it 27 times, most often “Brev: Start fine-tuning and training models in < 10 minutes”. No feed was found, so read it on the site itself.
Too few sampled comments (9) to say how often HN recommends it.?
Why HN links it
HN commenters mostly cite brev.dev as a cloud GPU rental product or list it among alternatives. The blog posts linked here cover fine-tuning Mistral on cloud GPUs and simulating two-player games to find equilibria.
Summarised by Claude from a sample of 9 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
Brev.Dev Instances Console
Most-linked posts?
- /4 people · 4 mentions · first linked 2022-02-10
- Brev: Start fine-tuning and training models in < 10 minutes4 people · 4 mentions · first linked 2022-12-28
- Don't Let a Bad Abstraction Cost You 2 Years2 people · 2 mentions · first linked 2022-06-17
- Automatically setup Dreambooth on a GPU environment2 people · 2 mentions · first linked 2022-11-01
What Hacker News says?
> What about alternatives? Custom models? Apart from the Big 3 (in no particular order): - https://together.ai/ - https://replicate.com/ - https://anyscale.com/ - https://baseten.co/ - https://modal.com/ - https://banana.dev/ - https://runpod.io/ - https://bentoml.com/ - https://brev.dev/ - https://octo.ai/ - https://cerebrium.ai/ ...
Comment linking /
on “Fly.io has GPUs now”Submitted Don't Let a Bad Abstraction Cost You 2 Years as a story · 17 points
HN discussionThe approach I see used is axolotl with QLoRA using cloud GPUs which can be quite cheap. https://github.com/OpenAccess-AI-Collective/axolotl Someone from one of the cloud GPU vendors wrote a guide: https://brev.dev/blog/fine-tuning-mistral
Comment linking /blog/fine-tuning-mistral
on “Ask HN: Most efficient way to fine-tune an LLM in 2024?”Submitted A simple guide to fine tuning Llama 2 on your own data as a story · 15 points
HN discussionNice! I'll have to read through this and use it to improve https://brev.dev/blog/how-to-win-a-2-player-game-of-are-you-... which was a similar exercise on a game slightly different than prisoners dilemma. I like being able to use "brute force" numerical methods to solve problems like this because they are often simpler to construct and exploit the fact that computers go brrr.
Comment linking /blog/how-to-win-a-2-player-game-of-are-you-a-robot
on “Finding Nash equilibria through simulation”Submitted Our experience upgrading from go v1.17 to v1.18 for generics as a story · 14 points
HN discussionThis is a huge pain, mostly stemming from using the same computer for different projects, because dependencies start to conflict. Big companies (fb, google, workday) solve this with cloud-hosted dev environments. Cloud environments let me contribute to our code base as a technical founder without having to have the technical expertise of a CTO. I'd rather not go back because while I know how to code, I'd rather not…
Comment linking /
on “Ask HN: How do you find (again) joy in software dev with so much setup?”Submitted Run Stable Diffusion in 4 minutes as a story · 9 points
HN discussionI feel your pain, I try to mitigate this as much as possible with templating projects. I use https://gitlab.com/reedrichards/cookiecutter-goapp for small webservers and this generates the infra i need for aws, and go has been a relatively simple development and packaging platform for me. I am also working at https://brev.dev and have been using their product at work and on personal projects to help standardize my de…
Comment linking /
on “Ask HN: How do you find (again) joy in software dev with so much setup?”Submitted Our experience migrating to Golang v1.18 for generics as a story · 7 points
HN discussionYou should check out https://brev.dev . You can rent GPUs, pause instances, use your own AWS/GCP accounts to make use of your credits, and the CLI lets you use your GPU as if it’s on your local machine.
Comment linking /
on “Meta AI announces Massive Multilingual Speech code & models for 1000+ languages”Submitted Generate secret messages in images with ControlNet Stable Diffusion WebUI as a story · 5 points
HN discussionalso, https://brev.dev/
Comment linking /
on “OpenLLaMA: An Open Reproduction of LLaMA”Submitted How Much Does It Cost to Train a Large Language Model? A Guide as a story · 4 points
HN discussionI think there's https://brev.dev and https://banana.dev
Comment linking /
on “New AI edits images based on text instructions (instructPix2Pix/imaginAIry)”Submitted A simple guide to fine-tuning Llama 2 as a story · 4 points
HN discussionR.I.P. battery. Personally I've been using Brev [1] to do my cloud training, you get a cloud GPU instance that you can upgrade/downgrade on the fly, and makes supports VS Code out of the box. [1] https://brev.dev/ (I'm not affiliated with Brev)
Comment linking /
on “Tensorbook”Submitted Run Stable LM with Brev as a story · 3 points
HN discussion404 Link should be https://brev.dev/blog/the-no-bs-guide-to-fine-tuning-an-llm
Comment linking /blog/the-no-bs-guide-to-fine-tuning-an-llm
on “The no-BS guide to fine tuning an LLM”Submitted Using Brev to decrease our AWS instance costs by up to 94% as a story · 3 points
HN discussionSubmitted How to easily run Falcon-40B as a story · 2 points
HN discussionSubmitted Automatically setup Dreambooth on a GPU environment as a story · 2 points
HN discussionSubmitted Don't Let a Bad Abstraction Cost You 2 Years as a story · 2 points
HN discussionSubmitted Brev: Start fine-tuning and training models in < 10 minutes as a story · 1 points
HN discussionSubmitted How QLoRA works to fine-tune large models with small GPUs as a story · 1 points
HN discussionSubmitted The no-BS guide to fine tuning an LLM as a story · 1 points
HN discussionSubmitted Dreambooth cloud GPU environment free for 30 minutes as a story · 1 points
HN discussion
