Luca Soldaini 🎀
banner
soldaini.net
Luca Soldaini 🎀
@soldaini.net
I like tokens! Lead for OLMo data at @ai2.bsky.social (Dolma 🍇) w @kylelo.bsky.social. Open source is fun 🤖☕️🍕🏳️‍🌈 Opinions are sampled from my own stochastic parrot

more at https://soldaini.net
best commute on earth
September 17, 2025 at 3:08 PM
my keystrokes go though light-up starry cable

OF COURSE my code is better than yours
August 20, 2025 at 4:08 PM
12+ years in this country, first time I get to wear this sticker 🗳️
August 5, 2025 at 3:29 PM
new @ai2.bsky.social office has something for everyone: stunning views for the outdoorsy kind, 2.5 Gbps connection at every desk for the indoor nerds
June 23, 2025 at 10:07 PM
today might be rainy, but PNW summer is already here
May 31, 2025 at 11:32 PM
when someone says they wanna bring me to their favorite italian restaurant
April 22, 2025 at 1:33 AM
bluesky deserves to know we’ve adopted a cat and he’s the most handsome boy
March 26, 2025 at 4:07 AM
"excuse me sir do you have a moment to talk about olmOCR"
March 10, 2025 at 7:54 PM
in the upcoming LLMs war, i choose a neutral team
February 20, 2025 at 8:57 PM
ok but how many LLMs you know are available on BluRay
February 7, 2025 at 8:19 AM
it's gonna be a sad day when thinking models will figure out ASCII art, current capabilities crack me up so much:
January 27, 2025 at 2:32 AM
just learned how gorgeous the old Seattle Times building was, so sad it got demolished for another downtown tower
January 13, 2025 at 4:51 PM
All together, these four aspects help us achieve the best fully open model yet.

🚗read the full tech report: arxiv.org/abs/2501.00656

💬 try OLMo 2 Instruct 13B on the @allen_ai playground: playground.allenai.org
January 3, 2025 at 7:51 PM
🤖 Infrastructure

A lot of LM shops are shy when it comes to describing how their infrastructure works. Not us!

We describe our two clusters, Augusta and Jupiter, in great details, and explain what matters in making your infra reliable
January 3, 2025 at 7:51 PM
🐪 Post-training with Tulu 3

OLMo 2 post-training is proof that our Tulu 3 recipe is easy to adapt to any model. Took the team less than 6 days to get a strong OLMo 2-Instruct checkpoint out!
January 3, 2025 at 7:51 PM
🧑‍🍳 Mid-training recipe

LLM capabilities are unlocked at the end of pretraining by curating the right mixture to show models last. We explain how to curate the right data for this stage & measure its effects
January 3, 2025 at 7:51 PM
💪 Stability

How do you ensure that your pretrain run doesn't blow up?

We spent months perfecting our OLMo recipe---it takes many targeted mitigations and expensive experiments to get it right… well now you don't have to!
January 3, 2025 at 7:51 PM
OLMo 2 tech report is out!

We get in the weeds with this one, with 50+ pages on 4 crucial components of LLM development pipeline:
January 3, 2025 at 7:51 PM
looking for new color themes for my terminal and it's hard to pass on this one
December 30, 2024 at 6:12 PM
Good Camera™️ pics
December 26, 2024 at 1:27 AM
rough seas but it was all so worth it 🌊🚢🥰
December 26, 2024 at 1:18 AM
i love the internet
December 25, 2024 at 3:18 AM
Anthropic, OpenAI: mid-century revival
Meta, Google: post Corporate Memphis, big-tech minimalism

I'm ready to give my money to the first brutalist AI endeavor 😍
December 20, 2024 at 5:48 PM
oh no @bnewbold.net do you know why wayback machine struggles to capture Bluesky web app? any recommendations?
December 13, 2024 at 11:52 PM
fella keeps staring at me no matter how much i tell him in not interested in their AI startup
December 13, 2024 at 9:58 PM