Made a list of resources for open source language models with @soldaini.net ahead of the tutorial tomorrow at 930 AM.
github.com/allenai/awes...
github.com/allenai/awes...
GitHub - allenai/awesome-open-source-lms: Friends of OLMo and their links.
Friends of OLMo and their links. Contribute to allenai/awesome-open-source-lms development by creating an account on GitHub.
github.com
December 10, 2024 at 1:25 AM
Made a list of resources for open source language models with @soldaini.net ahead of the tutorial tomorrow at 930 AM.
github.com/allenai/awes...
github.com/allenai/awes...
We just updated the OLMo repo at github.com/allenai/OLMo!
There are now several training configs that together reproduce the training runs that lead to the final OLMo 2 models.
In particular, all the training data is available, tokenized and shuffled exactly as we trained on it!
There are now several training configs that together reproduce the training runs that lead to the final OLMo 2 models.
In particular, all the training data is available, tokenized and shuffled exactly as we trained on it!
GitHub - allenai/OLMo: Modeling, training, eval, and inference code for OLMo
Modeling, training, eval, and inference code for OLMo - allenai/OLMo
github.com
December 2, 2024 at 8:13 PM
We just updated the OLMo repo at github.com/allenai/OLMo!
There are now several training configs that together reproduce the training runs that lead to the final OLMo 2 models.
In particular, all the training data is available, tokenized and shuffled exactly as we trained on it!
There are now several training configs that together reproduce the training runs that lead to the final OLMo 2 models.
In particular, all the training data is available, tokenized and shuffled exactly as we trained on it!
olmocr est un outil permettant de linéariser les documents PDF en vue de leur exploitation par des LLM à des fins d’entraînement ou de création de datasets. Très pratique pour extraire de la donnée relativement structurée et exploitable ⬇️
github.com/allenai/olmocr
github.com/allenai/olmocr
GitHub - allenai/olmocr: Toolkit for linearizing PDFs for LLM datasets/training
Toolkit for linearizing PDFs for LLM datasets/training - allenai/olmocr
github.com
March 1, 2025 at 7:01 AM
olmocr est un outil permettant de linéariser les documents PDF en vue de leur exploitation par des LLM à des fins d’entraînement ou de création de datasets. Très pratique pour extraire de la donnée relativement structurée et exploitable ⬇️
github.com/allenai/olmocr
github.com/allenai/olmocr
AllenAI Wants Everyone Else to Train MoEs
https://uenozooo.com/allenai-wants-everyone-else-to-train-moes/
#HuggingFace #AI #OpenSource
https://uenozooo.com/allenai-wants-everyone-else-to-train-moes/
#HuggingFace #AI #OpenSource
October 1, 2026 at 8:00 PM
AllenAI Wants Everyone Else to Train MoEs
https://uenozooo.com/allenai-wants-everyone-else-to-train-moes/
#HuggingFace #AI #OpenSource
https://uenozooo.com/allenai-wants-everyone-else-to-train-moes/
#HuggingFace #AI #OpenSource
🔥 allenai/Llama-3.1-Tulu-3-8B (trained with PPO) -> allenai/Llama-3.1-Tulu-3.1-8B (trained with GRPO)
We are happy to "quietly" release our latest GRPO-trained Tulu 3.1 model, which is considerably better in MATH and GSM8K!
We are happy to "quietly" release our latest GRPO-trained Tulu 3.1 model, which is considerably better in MATH and GSM8K!
February 12, 2025 at 5:33 PM
🔥 allenai/Llama-3.1-Tulu-3-8B (trained with PPO) -> allenai/Llama-3.1-Tulu-3.1-8B (trained with GRPO)
We are happy to "quietly" release our latest GRPO-trained Tulu 3.1 model, which is considerably better in MATH and GSM8K!
We are happy to "quietly" release our latest GRPO-trained Tulu 3.1 model, which is considerably better in MATH and GSM8K!
Reading the TÜLU 3 paper from @ai2.bsky.social. It's refreshing to see a research lab treating AI as a real science with full reports, data, code, logs, evals.
Paper: allenai.org/papers/tulu-...
Demo: playground.allenai.org
Code: github.com/allenai/open...
Eval: github.com/allenai/olmes
Notes
Paper: allenai.org/papers/tulu-...
Demo: playground.allenai.org
Code: github.com/allenai/open...
Eval: github.com/allenai/olmes
Notes
allenai.org
November 24, 2024 at 6:04 PM
Reading the TÜLU 3 paper from @ai2.bsky.social. It's refreshing to see a research lab treating AI as a real science with full reports, data, code, logs, evals.
Paper: allenai.org/papers/tulu-...
Demo: playground.allenai.org
Code: github.com/allenai/open...
Eval: github.com/allenai/olmes
Notes
Paper: allenai.org/papers/tulu-...
Demo: playground.allenai.org
Code: github.com/allenai/open...
Eval: github.com/allenai/olmes
Notes
These leaks are getting crazy. First the Claude Code source code leaks, and now apparently someone from AI2 leaked the olmo training code on github?
github.com/allenai/olmo...
github.com/allenai/olmo...
GitHub - allenai/OLMo-core: PyTorch building blocks for the OLMo ecosystem
PyTorch building blocks for the OLMo ecosystem. Contribute to allenai/OLMo-core development by creating an account on GitHub.
github.com
March 31, 2026 at 7:34 PM
These leaks are getting crazy. First the Claude Code source code leaks, and now apparently someone from AI2 leaked the olmo training code on github?
github.com/allenai/olmo...
github.com/allenai/olmo...
My response to ASU + OpenAI is that universities cannot afford to outsource aspects of *thinking*. We should partner instead with AllenAI & HuggingFace &c and help build public open-source models to support instruction.
January 19, 2024 at 5:38 AM
My response to ASU + OpenAI is that universities cannot afford to outsource aspects of *thinking*. We should partner instead with AllenAI & HuggingFace &c and help build public open-source models to support instruction.
Training data: huggingface.co/collections/...
Training code: github.com/allenai/open...
Eval code: github.com/allenai/olmes
Training code: github.com/allenai/open...
Eval code: github.com/allenai/olmes
November 21, 2024 at 5:15 PM
Training data: huggingface.co/collections/...
Training code: github.com/allenai/open...
Eval code: github.com/allenai/olmes
Training code: github.com/allenai/open...
Eval code: github.com/allenai/olmes
AllenAI just made trillion-parameter MoE training 2.7x faster with a clean architectural shift: keep experts GPU-resident, route data to them. Open source and benchmarked at 1.2T params on 512 B300 GPUs. https://huggingface.co/blog/allenai/olmocore3
October 2, 2026 at 6:04 AM
AllenAI just made trillion-parameter MoE training 2.7x faster with a clean architectural shift: keep experts GPU-resident, route data to them. Open source and benchmarked at 1.2T params on 512 B300 GPUs. https://huggingface.co/blog/allenai/olmocore3
Run olmOCR-bench yourself: github.com/allenai/olmo...
OCR your own documents: github.com/allenai/olmocr
Try the olmOCR online demo: olmocr.allenai.org
Read our updated technical report: olmocr.allenai.org/papers/olmoc...
OCR your own documents: github.com/allenai/olmocr
Try the olmOCR online demo: olmocr.allenai.org
Read our updated technical report: olmocr.allenai.org/papers/olmoc...
olmocr/olmocr/bench at main · allenai/olmocr
Toolkit for linearizing PDFs for LLM datasets/training - allenai/olmocr
github.com
June 18, 2025 at 3:23 PM
Run olmOCR-bench yourself: github.com/allenai/olmo...
OCR your own documents: github.com/allenai/olmocr
Try the olmOCR online demo: olmocr.allenai.org
Read our updated technical report: olmocr.allenai.org/papers/olmoc...
OCR your own documents: github.com/allenai/olmocr
Try the olmOCR online demo: olmocr.allenai.org
Read our updated technical report: olmocr.allenai.org/papers/olmoc...
Inspiring work! Btw your emo collection doesn’t exist on huggingface so the links to the models are broken (here and in your blog post).
For anyone else reading, this is the link for what appears to be the main model: huggingface.co/allenai/Emo_...
For anyone else reading, this is the link for what appears to be the main model: huggingface.co/allenai/Emo_...
allenai/Emo_1b14b_1T · Hugging Face
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
huggingface.co
May 8, 2026 at 4:56 PM
Inspiring work! Btw your emo collection doesn’t exist on huggingface so the links to the models are broken (here and in your blog post).
For anyone else reading, this is the link for what appears to be the main model: huggingface.co/allenai/Emo_...
For anyone else reading, this is the link for what appears to be the main model: huggingface.co/allenai/Emo_...
So happy OLMo 2 is out! We applied the same Tülu 3 RLVR recipe and it worked very nicely for our final 13B instruct model.
Here are the gains/losses of allenai/OLMo-2-1124-13B-Instruct (RLVR's checkpoint) over allenai/OLMo-2-1124-13B-DPO. More to share soon!
Here are the gains/losses of allenai/OLMo-2-1124-13B-Instruct (RLVR's checkpoint) over allenai/OLMo-2-1124-13B-DPO. More to share soon!
November 26, 2024 at 11:19 PM
So happy OLMo 2 is out! We applied the same Tülu 3 RLVR recipe and it worked very nicely for our final 13B instruct model.
Here are the gains/losses of allenai/OLMo-2-1124-13B-Instruct (RLVR's checkpoint) over allenai/OLMo-2-1124-13B-DPO. More to share soon!
Here are the gains/losses of allenai/OLMo-2-1124-13B-Instruct (RLVR's checkpoint) over allenai/OLMo-2-1124-13B-DPO. More to share soon!
Against the constant pressure of *genAI, genAI, genAI*, I am really appreciating @ai2.bsky.social 's work on creating tools for critical needs -- like crop maps and forest loss analysis. They just did a nice release on @hf.co.
huggingface.co/blog/allenai...
huggingface.co/blog/allenai...
OlmoEarth v1.1: A more efficient family of models
A Blog post by Ai2 on Hugging Face
huggingface.co
May 19, 2026 at 8:10 PM
Against the constant pressure of *genAI, genAI, genAI*, I am really appreciating @ai2.bsky.social 's work on creating tools for critical needs -- like crop maps and forest loss analysis. They just did a nice release on @hf.co.
huggingface.co/blog/allenai...
huggingface.co/blog/allenai...
𝗔𝗖𝗘𝟮𝗦-𝗦𝗛𝗶𝗘𝗟𝗗+, our new climate emulator that learns to separate the effects of sea surface temperature & CO2, is now on @hf.co—check it out → huggingface.co/allenai/ACE2...
June 10, 2026 at 5:46 PM
𝗔𝗖𝗘𝟮𝗦-𝗦𝗛𝗶𝗘𝗟𝗗+, our new climate emulator that learns to separate the effects of sea surface temperature & CO2, is now on @hf.co—check it out → huggingface.co/allenai/ACE2...
@BerkeleyEECS, @SimonsInstituteTOC, @LTIatCMU, @FARAIResearch, @SFIScience, @allenai, @srush_nlp, @PrincetonPLI, @SimonsFoundation, @openuk_uk, @USCISI, @NewHorizonsLanguageScience, @uwcse
January 5, 2025 at 4:11 AM
@BerkeleyEECS, @SimonsInstituteTOC, @LTIatCMU, @FARAIResearch, @SFIScience, @allenai, @srush_nlp, @PrincetonPLI, @SimonsFoundation, @openuk_uk, @USCISI, @NewHorizonsLanguageScience, @uwcse
Need AllenAI or someone like that to train a very small model for just this purpose
making world's most overengineered thermostat by explaining my air quality preferences to a local LLM and tasking it with regularly monitoring all the sensors and figuring out how to get the room as close as possible to the desired state
February 23, 2026 at 3:09 PM
Need AllenAI or someone like that to train a very small model for just this purpose
I'm really happy with Tulu 3 8B. Is nice in the hosted demo (playground allenai org), but also quantized running locally on my mac in LM studio D. Feels like a keeper.
November 22, 2024 at 8:55 PM
I'm really happy with Tulu 3 8B. Is nice in the hosted demo (playground allenai org), but also quantized running locally on my mac in LM studio D. Feels like a keeper.
October 2, 2026 at 12:22 AM
What's with these arguments over whether X or Y or whatever was the first LLM RL library? These all came in the last 3 months
We wrote multiturn RL4LMs like 3+ years ago github.com/allenai/RL4LMs
There were other simple versions even before. ML ppl approaching goldfish memory
We wrote multiturn RL4LMs like 3+ years ago github.com/allenai/RL4LMs
There were other simple versions even before. ML ppl approaching goldfish memory
GitHub - allenai/RL4LMs: A modular RL library to fine-tune language models to human preferences
A modular RL library to fine-tune language models to human preferences - allenai/RL4LMs
github.com
May 7, 2025 at 3:20 PM
What's with these arguments over whether X or Y or whatever was the first LLM RL library? These all came in the last 3 months
We wrote multiturn RL4LMs like 3+ years ago github.com/allenai/RL4LMs
There were other simple versions even before. ML ppl approaching goldfish memory
We wrote multiturn RL4LMs like 3+ years ago github.com/allenai/RL4LMs
There were other simple versions even before. ML ppl approaching goldfish memory
To gather feedback, we're releasing de-identified annotated tutoring transcripts plus code & scored replays.
💻 Code: buff.ly/gWlpQoR
🤗 Data: buff.ly/cEFVWWD
🌐 Learn more: buff.ly/yNT1lwP
💻 Code: buff.ly/gWlpQoR
🤗 Data: buff.ly/cEFVWWD
🌐 Learn more: buff.ly/yNT1lwP
GitHub - allenai/tutormoments
Contribute to allenai/tutormoments development by creating an account on GitHub.
github.com
August 7, 2026 at 4:26 PM
To gather feedback, we're releasing de-identified annotated tutoring transcripts plus code & scored replays.
💻 Code: buff.ly/gWlpQoR
🤗 Data: buff.ly/cEFVWWD
🌐 Learn more: buff.ly/yNT1lwP
💻 Code: buff.ly/gWlpQoR
🤗 Data: buff.ly/cEFVWWD
🌐 Learn more: buff.ly/yNT1lwP
Olmo 3 32B Think is now free on Open Router until 12/22.
Let us know what you think and what to improve!
(Hosted by Parasail)
This may give it the hug of death... would be my dream.
openrouter.ai/allenai/olmo...
Let us know what you think and what to improve!
(Hosted by Parasail)
This may give it the hug of death... would be my dream.
openrouter.ai/allenai/olmo...
December 5, 2025 at 12:49 AM
Olmo 3 32B Think is now free on Open Router until 12/22.
Let us know what you think and what to improve!
(Hosted by Parasail)
This may give it the hug of death... would be my dream.
openrouter.ai/allenai/olmo...
Let us know what you think and what to improve!
(Hosted by Parasail)
This may give it the hug of death... would be my dream.
openrouter.ai/allenai/olmo...