OLMo Models by Allen Institute for Artificial Intelligence (AI2)
About
The OLMo family of large language models (LLMs) is a series of open-source models developed by the Allen Institute for Artificial Intelligence (AI2) to advance the science of language modeling. These models stand out for their openness, offering researchers access to the training data, code, models, and evaluation resources. This transparency helps examine various aspects of LLM development, including biases and risks. Trained on the Dolma dataset, OLMo models utilize the Tulu SFT mixture and a refined UltraFeedback dataset for enhanced question answering. The family includes different models with varying parameters, such as 1B and 7B, reflecting distinct training phases and optimizations, and encourages collaborative research within the open-source AI community 123.
Current Variants
Use-when guidance is derived from seed capabilities, context, release, and replacement fields.
Use when the workload needs 64k context, 32B parameters, and reasoning.
Use when the workload needs 64k context, 32B parameters, and tool use.
Use when the workload needs 7B parameters and structured outputs.
Use when the workload needs 1B parameters and structured outputs.
Use when the workload needs 7B parameters and structured outputs.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| OLMo 3 32B Think | Use when the workload needs 64k context, 32B parameters, and reasoning. | 2026-03 | 64k context32B parametersreasoning | Current |
| OLMo 3.1 32B Instruct | Use when the workload needs 64k context, 32B parameters, and tool use. | 2026-02 | 64k context32B parameterstool use | Current |
| OLMo 7B | Use when the workload needs 7B parameters and structured outputs. | 2024-02 | 7B parametersstructured outputs | Current |
| OLMo 1B | Use when the workload needs 1B parameters and structured outputs. | 2024-02 | 1B parametersstructured outputs | Current |
| OLMo 7B Twin-2T | Use when the workload needs 7B parameters and structured outputs. | 2024-02 | 7B parametersstructured outputs | Current |
| OLMo 1.7 7B | Use when the workload needs 7B parameters. | 2024-02 | 7B parameters | Current |
Release Timeline
3 release groupsSpecifications(6 models)
| Model | Released | Context | Parameters | Reasoning | Fn Calling | Tool Use | Structured Outputs |
|---|---|---|---|---|---|---|---|
| OLMo 3 32B Think | 2026-03 | 64k | 32B | Yes | No | No | No |
| OLMo 3.1 32B Instruct | 2026-02 | 64k | 32B | No | Yes | Yes | Yes |
| OLMo 7B | 2024-02 | — | 7B | No | No | No | Yes |
| OLMo 1B | 2024-02 | — | 1B | No | No | No | Yes |
| OLMo 7B Twin-2T | 2024-02 | — | 7B | No | No | No | Yes |
| OLMo 1.7 7B | 2024-02 | — | 7B | No | No | No | No |
Available From(3 providers)
Pricing
| Model | Provider | Input / 1M | Output / 1M | Type |
|---|---|---|---|---|
| OLMo 7B | Together AI | $0.2 | $0.2 | Serverless |
| OLMo 7B Twin-2T | Together AI | $0.2 | $0.2 | Serverless |
| OLMo 1B | OpenRouter | $15 | $60 | Serverless |
Frequently Asked Questions
- What is OLMo used for?
- OLMo is used for reasoning, agent workflows and tool use, and structured outputs. The family description and listed model capabilities point to those workloads as the best fit.
- How does OLMo compare to Tulu?
- OLMo by Allen Institute for Artificial Intelligence (AI2) is strongest where you need reasoning, while Tulu by Allen Institute for Artificial Intelligence (AI2) is the closest related family to check for adjacent model selection. OLMo has 6 listed variants and reaches up to 64k context, while Tulu reaches up to 4k context, so compare the specs and pricing tables before choosing a production model.
- Which OLMo model should I use?
- For the lowest listed input price, start with OLMo 7B through Together AI at $0.2/1M input tokens. For the most capable/latest local choice, evaluate OLMo 3.1 32B Instruct with 64k context and tool use, function calling, and structured outputs.
