TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Show HN: Exafunction NLP – Run models like GPT-J for 6x cheaper than OpenAI

45 pointsby varunkmohanover 3 years ago

9 comments

varunkmohanover 3 years ago
Hi HN, I’m Varun, cofounder of Exafunction. We’re excited about all the use cases of large NLP models like text generation and language comprehension, but found the existing offerings to be quite expensive. Our goal is to serve these models in a cost effective way at scale.<p>The first model we’re starting with is GPT-J, the recently released open source large language model by EleutherAI. It’s comparable to OpenAI’s GPT-3 Curie in performance. With lots of optimizations, we are able to serve the model for 6x cheaper per token compared to OpenAI.<p>We have a simple HTTPS API that you can try out for free. Here’s a link with the details - <a href="https:&#x2F;&#x2F;www.exafunction.com&#x2F;nlp-api" rel="nofollow">https:&#x2F;&#x2F;www.exafunction.com&#x2F;nlp-api</a><p>We’re also looking into supporting custom and fine-tuned models since it’s required to get good performance for a lot of applications. Especially for smaller models like BERT, we’re excited about the cost savings we can deliver for users who use many different models. In this case, we’re 100 - 1000x cheaper than others offering similar inference APIs like Hugging Face.<p>We hope this is interesting to you all and please email me at varun@exafunction.com if there’s a model you’d like to see supported.
评论 #28691069 未加载
评论 #28688312 未加载
axb1368over 3 years ago
Have you considered adding support for CodeT5? I believe they&#x27;re supported in Huggingface as well. <a href="https:&#x2F;&#x2F;github.com&#x2F;salesforce&#x2F;CodeT5" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;salesforce&#x2F;CodeT5</a>
评论 #28688339 未加载
ml_hardwareover 3 years ago
Wow! How are you able to achieve the cost reductions? Is it different hardware, software optimizations, or both? Also does this suggest that OpenAI is charging 6x markups on their models... &gt;:(
评论 #28686441 未加载
LuisMondragonover 3 years ago
Awesome. I am testing HuggingFace&#x27;s API and evaluating if I should pay for a dev account. I will evaluate Exafunction also, it looks interesting. Specifically, I&#x27;m interested in using the multilingual LaBSE model. Do you support this model?
评论 #28686188 未加载
suryabhupaover 3 years ago
This is really remarkable! How hard do you think it will be to support new models, i.e. does the tooling you’ve built generalize to you being able to serve other large scale models easily?
ajaysaini235over 3 years ago
Wow this is really cool! How did you guys get such big speedups?
评论 #28686502 未加载
lxyover 3 years ago
Cool! 7500 words free per month…does anyone have any interesting ideas about what you could do with this?
pqnover 3 years ago
I have a couple of NLP side projects that I&#x27;m working on and I&#x27;ll definitely try using this!
abistarostaover 3 years ago
amazing cost ... will try it out for some personal projects