olmo
13B parameters
Commercial OK
Reviewed May 2026OLMo 2 13B
AI2's fully-open 13B. Apache 2.0; full training data + checkpoints + recipes published. The reproducibility-first model in the 13B class.
License: Apache 2.0·Released Nov 26, 2024·Context: 4,096 tokens
Overview
AI2's fully-open 13B. Apache 2.0; full training data + checkpoints + recipes published. The reproducibility-first model in the 13B class.
Strengths
- Fully-open training data
- Apache 2.0
- AI2 research backing
Weaknesses
- Trails Qwen 2.5 / Llama 3.x on benchmarks
Quantization variants
Each quantization trades model quality for file size and VRAM. Q4_K_M is the most popular starting point.
| Quantization | File size | VRAM required |
|---|---|---|
| Q4_K_M | 7.8 GB | 10 GB |
Get the model
HuggingFace
Original weights
huggingface.co/allenai/OLMo-2-1124-13B-Instruct
Source repository — direct quantization required.
Hardware that runs this
Cards with enough VRAM for at least one quantization of OLMo 2 13B.
Frequently asked
What's the minimum VRAM to run OLMo 2 13B?
10GB of VRAM is enough to run OLMo 2 13B at the Q4_K_M quantization (file size 7.8 GB). Higher-quality quantizations need more.
Can I use OLMo 2 13B commercially?
Yes — OLMo 2 13B ships under the Apache 2.0, which permits commercial use. Always read the license text before deployment.
What's the context length of OLMo 2 13B?
OLMo 2 13B supports a context window of 4,096 tokens (about 4K).
Source: huggingface.co/allenai/OLMo-2-1124-13B-Instruct
Reviewed by RunLocalAI Editorial. See our editorial policy for how we research and verify model claims.
Related — keep moving
Compare hardware
Buyer guides
Before you buy
Verify OLMo 2 13B runs on your specific hardware before committing money.