Haven (YC S23) – Quickly iterate when fine-tuning open-source LLMs

Hacker News

Demo Video: https://www.youtube.com/watch?v=XpyVKUyt7k8 Hey all! A friend and I have been building projects with open-source LLMs for a while now (originally for other project ideas) and found that quickly iterating with different fine-tuning datasets is super hard. Training a model, setting up some inference code to try out the model and then going back and forth took 90% of our time. That’s why we built Haven, a service to quickly try out different fine-tuning datasets and base-models. Going from uploading a dataset to chatting with the resulting model now takes less than 5 minutes (using a reasonably sized dataset). We fine-tune the models using low-rank adapters, which not only means that the changes made to the model are very small (only 30mb for a 7b parameter LLM), it also allows us to host many fine-tuned models very efficiently by hot swapping adapters on demand. This helped us reduce cold-start times to below one second and makes it possible for us to host a single trained model for a few dollars per month. [Research has shown]( https://arxiv.org/pdf/2305.14314.pdf ) that low-rank fine-tuning performance stays almost on-par with full fine-tuning. We charge between $0.004/1k training tokens, and after signing up, you get $5 in free credits. You can export all the models to Huggingface. Right now we support Llama-2 and Zephyr (which is itself a fine-tune of Mistral) as base models. We’re gonna add some more soon. We hope you find this useful and we would love your feedback!

Visit product View on Hacker NewsNov 28, 2023

Share card

Actual performance

13points

1comments

Made the leaderboard

Launch Intel predictions

Analyze your own launch →

Product HuntOn track for Day 1 leaderboard · Strong signals: model, models, single · Missing: mac, agents, macos

90%90% predicted probability of success on Product Hunt, based on ML models trained on real launch data.

best fitHighest predicted score across all platforms for this description.

Indie HackersFits the IH revenue-focused audience · Strong signals: para, efficiently · Missing: supports, reddit linkedin, podcasting

84%84% predicted probability of success on Indie Hackers, based on ML models trained on real launch data.

Hacker NewsStrong engagement from HN community · Strong signals: llama, ide · Missing: https docs, excited, just released

72%72% predicted probability of success on Hacker News, based on ML models trained on real launch data.

nativeThis product was originally launched on this platform.

AppSumoMay struggle as an AppSumo deal · Strong signals: host, soon, efficient · Missing: plus, platform, intuitive

48%48% predicted probability of success on AppSumo, based on ML models trained on real launch data.

TrustMRRLess likely to generate early MRR · Strong signals: video, month, para · Missing: mobile apps, ios, personal

42%42% predicted probability of success on TrustMRR, based on ML models trained on real launch data.

Acquire.comPre-revenue stage for this audience · Strong signals: training · Missing: arr, mrr, revenue

17%17% predicted probability of success on Acquire.com, based on ML models trained on real launch data.

BetaListMay not resonate with beta-testers · Strong signals: chat · Missing: web3, crypto, cryptocurrency

0%0% predicted probability of success on BetaList, based on ML models trained on real launch data.

Correct prediction on native model

Similar products

Open Source Reinforcement Fine-Tuning for Your Agents49%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

Open Source Reinforcement Fine-Tuning for Your Agents

Hacker News5

Terracotta – Platform for fine-tuning and evaluating LLMs40%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

Terracotta – Platform for fine-tuning and evaluating LLMs

Hacker News1

Open-source fine-tuning in a Colab notebook80%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

Open-source fine-tuning in a Colab notebook

Hacker News5

Fine-Tuning Index of Open-Source LLMs vs. OpenAI58%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

Fine-Tuning Index of Open-Source LLMs vs. OpenAI

Hacker News4

Fine tuning and RLHF mistralai 7B using DeepSpeed38%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

Fine tuning and RLHF mistralai 7B using DeepSpeed

Hacker News1

Courtyard – Open-source macOS app for local MLX fine-tuning Text52%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

Courtyard – Open-source macOS app for local MLX fine-tuning Text

Hacker News1

ShadowPEFT – Centralized and Detachable Parameter-Efficient Fine-Tuning44%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

ShadowPEFT – Centralized and Detachable Parameter-Efficient Fine-Tuning

Hacker News6

FinetuneDB59%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

AI fine-tuning platform to create custom LLMs

Product Hunt+52

Predibase Reinforcement Fine-Tuning60%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

LLM reinforcement fine-tuning platform to improve LLM output

Product Hunt+172SaaS

Quaterion – x100 faster fine-tuning of similarity learning models49%Launch Intel prediction score: how likely this product is to succeed on its source platform, based on its name, tagline, and description.

Quaterion – x100 faster fine-tuning of similarity learning models

Hacker News19

Browse all products Analyze your launch