OpenPipe @OpenPipeAI

OpenPipe: Fine-tuning for production apps. Train higher quality, faster models. (YC S23) openpipe.ai San Francisco Joined August 2023

Tweets

93
Followers

4K
Following

4
Likes

108

Julia Neagu @JuliaANeagu

12 months ago

I write about this shift and the tools I expect to be part of the permanent core dev stack: @togethercompute, @FireworksAI_HQ , @GroqInc, @OpenPipeAI, @withmartian, @UnstructuredIO, @LangChainAI , @llama_index , @pinecone and @qdrant_engine. Check out the article here:…

2 3 24 14K 46

Andy Ayrey @AndyAyrey

12 months ago

@valeeaio @truth_terminal Yep fine tune on llama using @OpenPipeAI

2 2 12 13K 9

We saw lots of fun games at our AIxGames Meetup with Clementine & @OpenPipeAI. The Groq ASR API provides ultra-low latency audio transcription & translation for Whisper models. If you're a game developer, reach out to [email protected] to discuss the best model for your needs.

0 5 15 11K 4

Download Image

OpenPipe @OpenPipeAI

11 months ago

OpenPipe linked up w/ Wyatt Marshall CTO & Co-Founder of Halluminate so he could have an in-depth conversation on how to build a robust Evals system for your production GenAI technology w/ Reid Mayo (Founding AI Engineer). Check it out!: youtube.com/watch?v=1ygD4o…

0 5 20 5K 19

Joshua Schrier @JoshuaSchrier

a year ago

@andrewwhite01 @OpenPipeAI is super easy to use.

2 1 3 4K 3

Kyle Corbitt @corbtt

a year ago

@WisprAI gets very low latency, low costs and extremely high quality by using fine-tuned models through @OpenPipeAI. Proud to have them as a customer, and congrats on the launch!

Wispr Flow @WisprFlow

a year ago

@WisprAI gets very low latency, low costs and extremely high quality by using fine-tuned models through @OpenPipeAI. Proud to have them as a customer, and congrats on the launch!

161 174 1K 1.7M 1K

Download Video

1 1 5 5K 4

OpenPipe @OpenPipeAI

a year ago

Our CEO @corbtt is giving a talk next Wed in SF w/ @neutralino1 and @charles_irl. Kyle's topic is on fine-tuning best practices to get far higher quality LLM results vs prompting alone, and when to explore this optimization technique. check it out! lu.ma/jfsyz0xc

0 2 12 3K 8

Download Image

OpenPipe @OpenPipeAI

a year ago

We <3 Helicone

Helicone @helicone_ai

a year ago

We <3 Helicone

0 1 6 3K 12

0 2 9 3K 3

OpenPipe @OpenPipeAI

a year ago

I am not sure whether we want more Sydneys in the world but if you do you can 100% use OpenPipe to build them. 🫡

GwynTel™ @gwyntel

a year ago

I am not sure whether we want more Sydneys in the world but if you do you can 100% use OpenPipe to build them. 🫡

0 0 1 2K 1

Download Image

0 2 2 2K 2

OpenPipe @OpenPipeAI

a year ago

PSA because apparently we're getting confused with closed-down competitors: all OpenPipe users can export their fine-tuned models at any time. No lock-in nonsense here!

Tim Carambat @tcarambat

a year ago

PSA because apparently we're getting confused with closed-down competitors: all OpenPipe users can export their fine-tuned models at any time. No lock-in nonsense here!

1 0 3 1K 0

0 0 7 1K 3

Mike Bird @MikeBirdTech

a year ago

@FieroTy Services used: - @OpenPipeAI - @GroqInc - @firecrawl_dev - @ExaAILabs - @NousResearch - @AnthropicAI - @arxiv

0 2 11 2K 10

KALALA NZENIELE @cniongolo

a year ago

Over a year ago, I started @ClarityClips. Now, at @ClarityClips we've gained in automony and headed towards an open-source model implementation and a more solid sack. Thanks to @OpenPipeAI for their great service and credits

0 1 2 1K 1

Tejas Kumar @TejasKumar_

a year ago

NEW on ConTejas Code @corbtt, CEO @OpenPipeAI: How to fine-tune your own language models Through this discussion, I learned a ton from Kyle about machine learning, overfitting, hyperparameters, and a lot more. I hope you do too. Links in reply. x.com/TejasKumar_/st…

Tejas Kumar @TejasKumar_

a year ago

2 0 9 15K 5

Download Video

2 5 13 16K 12

Download Video

Finn Brown @finnatsea

a year ago

Most AI dev tools are solving fake problems. They are made by people who haven't done ML and sold to business customers who don't know what they want. @OpenPipeAI is a notable exception. Great product solving a real pain point.

1 2 11 2K 3

Joshua Schrier @JoshuaSchrier

a year ago

tl;dr—As a #GPUpoor man, I really enjoyed using @OpenPipeAI to facilitate #finetuning experiments across different models (and for many practical #chemistry problems you don't need to get into the weeds to accomplish the science goals.)

0 1 4 1K 2

Kyle Corbitt @corbtt

a year ago

I can't confirm or deny any specific model releases this week but just gonna say that if a SOTA open source model WERE to be released this week, @OpenPipeAI will have same-day support for fine-tuning.

Teknium (e/λ) @Teknium1

a year ago

70 133 1K 184K 201

Download Image

7 3 101 19K 29

Kyle Corbitt @corbtt

a year ago

If you ever felt the need for an Extremely Overpowered fine-tuned model... we now support training GPT-4o in @OpenPipeAI. Please use responsibly. 😎

8 17 183 29K 73

Download Image

Kyle Corbitt @corbtt

a year ago

Fantastic blog series from @strickvl who found his fine-tuned @OpenPipeAI model outperformed GPT-4 (as well as other fine-tuned models!) on response quality for a tiny fraction of the cost. Love it when our users share their success. 🤩 mlops.systems/posts/2024-07-…

1 6 37 4K 30

Wordware @wordware

a year ago

@corbtt and @saumyagandhi007 recently shared some insightful knowledge on optimizing outputs of LLMs with Mixtures of Agents (MoA) in their article on @OpenPipeAI👀 They introduced the 'Mixture of Agents' (MoA) models, which outperformed GPT-4 variants in real-world tasks,…

3 1 7 2K 2

Download Image

Harrison Jackson @HarrisonJackson

a year ago

@SullyOmarr @corbtt @OpenPipeAI @aidan_mclau Would recommend! They make the engineering effort easy enough you can basically ignore it and get a fine tuned model to test/run evals/actually use for < $50 of the training cost.