Allen Institute for AI · olmo2
Can I run olmo2?
OLMo 2 is a new family of 7B and 13B models trained on up to 5T tokens. These models are on par with or better than equivalently sized fully open models, and competitive with open-weight models such as Llama 3.1 on English academic benchmarks.
Parameter counts and memory belong to each build below. A family can contain several sizes and tasks; fitting in memory is not a quality ranking.
Calculate memory for a build and contextolmo2 system requirements
Memory needed at 4K context, counting model weights, KV cache, compute buffers and runtime overhead. Download sizes come from the official Ollama registry.
| Build | Quantization | Download | Memory needed |
|---|---|---|---|
| olmo2:7b7B parameters (tag)4K documented maximum contextBuild sourceCatalog checked 2026-10-13 | Q4_K_M | 4.2 GB | ~6.9 GB |
| olmo2:7b-1124-instruct-q8_07B parameters (tag)4K documented maximum contextBuild sourceCatalog checked 2026-10-13 | Q8_0 | 7.2 GB | ~10.2 GB |
| olmo2:13b13B parameters (tag)4K documented maximum contextBuild sourceCatalog checked 2026-10-13 | Q4_K_M | 7.8 GB | ~11.9 GB |
| olmo2:13b-1124-instruct-q8_013B parameters (tag)4K documented maximum contextBuild sourceCatalog checked 2026-10-13 | Q8_0 | 13.6 GB | ~18.2 GB |
| olmo2:7b-1124-instruct-fp167B parameters (tag)4K documented maximum contextBuild sourceCatalog checked 2026-10-13 | F16 | 13.6 GB | ~17.1 GB |
| olmo2:13b-1124-instruct-fp1613B parameters (tag)4K documented maximum contextBuild sourceCatalog checked 2026-10-13 | F16 | 25.6 GB | ~31.1 GB |
Showing the 6 smallest non-alias builds. Choose any build and change the context. A context above the build's documented maximum is a hypothetical calculation, not a supported setting.
How to run olmo2 locally
- 1.Install Ollama for Windows, macOS or Linux.
- 2.Download this exact build in a terminal:
ollama pull olmo2:7b - 3.Check the source for the build's task. For text generation, run
ollama run olmo2:7b. Base/text builds may need a prompt template; an instruct build is usually the appropriate starting point for a conversation.
Frequently asked questions about olmo2
- How much VRAM do I need to run olmo2?
- At 4K context, the smallest published build of olmo2 (Q4_K_M) needs roughly 6.9 GB of GPU memory once weights, KV cache, compute buffers and runtime overhead are counted. Bigger quantizations and longer context windows need more. It can also run partly or entirely on the CPU using system RAM, more slowly.
- Can I run olmo2 without a dedicated GPU?
- A supported CPU runtime can use system RAM. The smallest listed build needs an estimated 6.9 GB at 4K context, plus headroom for your operating system. Check the build's runtime requirements. We have not measured its speed on your CPU.
- How big is the olmo2 download?
- The smallest published build is 4.2 GB. There are 6 builds in total, the largest being 25.6 GB. Leave some extra free disk space beyond the download itself.
- How do I run olmo2 locally?
- Install Ollama, then download a specific build with "ollama pull olmo2:7b". For a build that supports text generation, use ollama run with the same tag. Check the model source for its task and prompt format.
Will olmo2 run on your PC?
Scan your hardware once and get a verdict for this model and every build — free, no account.
Compare with similar models
Or start from your memory: 8 GB VRAM, 12 GB VRAM, 16 GB VRAM, 24 GB VRAM or a 16 GB Mac.
Model data from the official Ollama registry · verified 13/10/2026