Organization
Ai2 (Allen Institute for AI)
Ai2 (Allen Institute for AI) is an AI model developer and research lab headquartered in Seattle, Washington12. This catalog lists 5 open model releases, 1 software project, and 1 dataset from Ai2 (Allen Institute for AI), under licenses including the Apache License 2.0 (Dolma toolkit and Dolma 3 construction code) and the Llama 3.1 Community License Agreement.
Documented roles: Model developer · Research lab · Nonprofit research
Ai2 is a non-profit AI research institute in Seattle, founded in 2014 by Paul Allen. Its work covers open models and training data, as well as robotics, conservation, and other applied research.1
Using this organization’s offerings
Intended users
Olmo is presented for research and development, with model-flow resources for studying and building language models.12
Hosted and self-hosted access
The Olmo page links model and training-data downloads, OlmoCore training tools, preprocessing tools and evaluation resources. Inspect the specific version linked in the catalog for its actual coverage.12
Restrictions and review limits
Availability across the project is not proof that every release has complete information or the same license. This profile links primary resources; a release-level rights and completeness audit remains separate.12
Key facts
- Legal name
- The Allen Institute for Artificial Intelligence3
- Ownership (separate from openness)
- Nonprofit
Ai2 describes itself as a nonprofit AI research institute. This organizational status is independent of the terms of each released artifact.11
- Legal form
- Corporation recognized by the IRS as tax-exempt under Section 501(c)(3); the IRS extract classifies it as a private operating foundation.34
- Founded
- 20141
- Sectors
- Open models, Research, Data
What is and isn’t public
Ai2's developer documentation states that it releases training code, training data, model weights, and recipes for its flagship models. This catalog assesses openness release by release (see the OLMo release records), not for the organization as a whole; Asta and Semantic Scholar are hosted services.5910
Products
Hosted model products
Labeled Hosted product/API — this describes how a documented product is delivered (hosted access). It does not determine whether underlying weights or code are open; assess the specific release and its licenses.
| Product | Kind | Documented delivery and access | Reviewed |
|---|---|---|---|
| Ai2 Playground (external site: playground.allenai.org) | Assistant app | Web chat interface for trying Ai2 models such as Olmo 3, including the OlmoTrace tool for tracing parts of a response back to training data.78 | Sep 29, 2026 |
| Asta (external site: asta.allen.ai) | Assistant app | Agentic research assistant for scholarly tasks, used on the web at asta.allen.ai; the Asta page also covers the AstaBench benchmark suite and developer resources and invites users to sign up for a preview.9 | Sep 29, 2026 |
Other products
| Product | Kind | Documented delivery and access | Reviewed |
|---|---|---|---|
| Semantic Scholar (external site: semanticscholar.org) | Other | Free AI-assisted search and discovery service for research literature, with the Semantic Scholar Academic Graph datasets and APIs for developers.10 | Sep 29, 2026 |
Cataloged artifacts
Model releases
- Llama 3.1 Tülu 3.1 8BModel releaseAvailability: PublicModel-disclosure tier (USASI rubric v0.2): Open-weight
- Molmo2-O 7BModel releaseAvailability: PublicModel-disclosure tier (USASI rubric v0.2): Open-weight
- Olmo 3 7B (base)Model releaseAvailability: PublicModel-disclosure tier (USASI rubric v0.2): Open-stack
- Olmo 3 7B InstructModel releaseAvailability: PublicModel-disclosure tier (USASI rubric v0.2): Open-stack
- Olmo 3.1 32B ThinkModel releaseAvailability: PublicModel-disclosure tier (USASI rubric v0.2): Open-stack
Frameworks
- OLMo-coreSoftwareAvailability: Public
Datasets
- DolmaDatasetAvailability: Public
Notable documented facts
- Ai2's developer documentation directs users who want hosted API access to Olmo models to OpenRouter and to Ai2's inference providers Cirrascale and Parasail.6
In the news
Ai2 releases HiRO-ACEv2, an updated climate emulator and 3 km downscaling system, under Apache 2.0
Ai2 announced HiRO-ACEv2 on October 14, 2026 as the newest update to its High Resolution Output Ai2 Climate Emulator system, pairing an updated ACE model with a new downscaling model, HiROv2. The model card describes HiROv2 as a diffusion model that generates 3 km precipitation, 10 m winds, and sea level pressure from coarse ACE output, and releases the ACE2S-X-SHiELD and HiROv2 checkpoints with inference configs under Apache 2.0 for research and educational use. The card says the model was not trained or tested outside 66S to 70N, is not expected to generalize to climates outside its 10-year training data, and that ACE2S-X-SHiELD is not suitable for weather forecasting. USASI has not tested this model.
Ai2's Bolmo byte-level language model research is published in Nature, with new checkpoints
Ai2 said on October 7, 2026 that the research behind Bolmo, its family of fully open byte-level language models introduced the previous December, has been published in Nature. Ai2 writes that the paper applies the same "byteifying" process to Qwen 3 8B and Llama 3 8B to create two new byte-level models, Bwen 8B and Blama 8B, and that it is releasing new checkpoints, including Stage 1 checkpoints, on Hugging Face. The Bwen 8B model card describes it as Qwen3 8B Base retrofitted to operate over bytes instead of tokens and lists the Apache 2.0 license. USASI has not tested these models.
Ai2 open-sources AstaBrief 8B, a model for writing cited scientific reports
On October 2, 2026 Ai2 announced that it is open-sourcing AstaBrief 8B, a model that turns a research question and retrieved literature excerpts into a cited report, along with its training data. Ai2 says the model was built from Qwen3-8B using supervised fine-tuning and direct preference optimization, and that it is available as Fast mode in the "Generate a report" feature of Asta, its platform for scientific work. The Hugging Face model card lists the Apache 2.0 license. Ai2 notes that most of the training and evaluation was completed in 2025 and that it has not rerun the full evaluation against today's frontier models.
Ai2 releases OLMo-core 3, its open training library, with mixture-of-experts support
Ai2 released version 3 of OLMo-core, the PyTorch training library behind its Olmo models, on October 1, 2026. The release redesigns the library's mixture-of-experts training system, adding expert and pipeline parallelism, a distributed optimizer, grouped matrix multiplication for experts, and support for the MXFP8 number format; Ai2 says it is designed to scale mixture-of-experts training into the trillion-parameter range. The code is licensed under Apache 2.0, and the repository includes Ai2's official training scripts for Olmo 3.
Ai2 releases BenchMIRT, a method, code, and data for auditing LLM benchmarks prompt by prompt
Ai2 introduced BenchMIRT on September 1, 2026, describing it as a method for auditing LLM benchmarks at the level of individual prompts. According to Ai2, it uses multidimensional Item Response Theory to estimate which underlying capabilities are most closely associated with getting each question right. Ai2 published a technical report, code on GitHub, and data in a Hugging Face collection. The post does not state a license. USASI has not tested this method.
U.S. eligibility
Eligible · basis: U.S. nonprofit or lab
Ai2 describes itself as a Seattle-based non-profit AI research institute, and its contact page lists a Seattle, Washington address. The IRS exempt-organization extract for Washington lists The Allen Institute for Artificial Intelligence in Seattle as a 501(c)(3) organization.123
Sources
This listing is not an endorsement or a federal approval.
Share or cite this page
- X (opens X)
- Bluesky (opens Bluesky)
- LinkedIn (opens LinkedIn)
- Reddit (opens Reddit)
- Hacker News (opens Hacker News)
- Email (opens your email app)
- RSS updates
Cite this page
APA style
USASI. (2026, October 1). Ai2 (Allen Institute for AI). United States of America Superintelligence. https://unitedstatesofamericasuperintelligence.com/companies/ai2/
BibTeX
@misc{usasi_ai2,
author = {{USASI}},
title = {Ai2 (Allen Institute for AI)},
year = {2026},
month = oct,
howpublished = {\url{https://unitedstatesofamericasuperintelligence.com/companies/ai2/}},
note = {United States of America Superintelligence. Last updated 2026-10-01}
}The date is when this page was last updated. Add the date you read it if your style needs one.
For maintainers: a “Listed in USASI” badge
Links to this page. It says the entry is listed, not endorsed or rated. About the badge
Markdown (for a README)
[](https://unitedstatesofamericasuperintelligence.com/companies/ai2/)
HTML
<a href="https://unitedstatesofamericasuperintelligence.com/companies/ai2/"><img src="https://unitedstatesofamericasuperintelligence.com/badges/listed.svg" alt="Listed in the USASI catalog" height="20"></a>