OLMo 2 — local AI model by Allen AI
Written by Jakub Rusinowski · Last updated
The world's most transparent LLM. OLMo 2 releases everything: weights, training code, training data, evaluation suite, and intermediate checkpoints. Perfect for researchers and compliance-sensitive deployments needing fully auditable AI. Competitive with Llama 3.1 on most benchmarks.
Variants
The smallest OLMo 2 variant needs about 5 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache.
| Model | VRAM at Q4 | VRAM | Context | Run it |
|---|---|---|---|---|
| OLMo 2 7B Instruct → 7B | ~5.2 GB | 4,096 | ollama run olmo2:7b | |
| OLMo 2 13B Instruct → 13B | ~9.1 GB | 4,096 | ollama run olmo2:13b |
Memory is quantized weights plus overhead at Q4_K_M, from the same engine as the GPU & VRAM checker.
How to run OLMo 2 locally
Install Ollama, then pull the tag.
ollama run olmo2:7bPick a size above for its own VRAM figure, speed estimate and install command.
Licence
Commercial use permitted. No usage restrictions beyond attribution.
Applies to: OLMo 2 7B Instruct, OLMo 2 13B InstructRecommended GPU
The cheapest catalogued GPU that runs OLMo 2 locally (min 5 GB VRAM) is the Intel Arc B570 (10 GB).
Can I run OLMo 2 on my GPU?
- OLMo 2 on AMD Radeon RX 7800 XT
- OLMo 2 on AMD Radeon RX 7900 XT
- OLMo 2 on AMD Radeon RX 9060 XT 16GB
- OLMo 2 on AMD Radeon RX 9060 XT 8GB
- OLMo 2 on AMD Radeon RX 9070
- OLMo 2 on AMD Radeon RX 9070 XT
- OLMo 2 on Intel Arc B570
- OLMo 2 on NVIDIA GeForce RTX 3070
- OLMo 2 on NVIDIA GeForce RTX 3070 Ti
- OLMo 2 on NVIDIA GeForce RTX 3080 (10GB)
- OLMo 2 on NVIDIA GeForce RTX 4060
- OLMo 2 on NVIDIA GeForce RTX 4060 Ti 16GB
- OLMo 2 on NVIDIA GeForce RTX 4070 Ti Super
- OLMo 2 on NVIDIA GeForce RTX 4080
- OLMo 2 on NVIDIA GeForce RTX 4080 Super
- OLMo 2 on NVIDIA GeForce RTX 5060
- OLMo 2 on NVIDIA GeForce RTX 5060 Ti 16GB
- OLMo 2 on NVIDIA GeForce RTX 5060 Ti 8GB
- OLMo 2 on NVIDIA GeForce RTX 5070 Ti
- OLMo 2 on NVIDIA GeForce RTX 5080
OLMo 2 — frequently asked questions
How much VRAM does OLMo 2 need?
OLMo 2 needs about 5 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: OLMo 2 7B Instruct (5 GB, Q4_K_M); OLMo 2 13B Instruct (9 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Can I run OLMo 2 on an RTX 4090 (24 GB)?
Yes — OLMo 2 runs on an RTX 4090 (24 GB) and other 24 GB cards such as the RTX 3090. Smaller variants also fit comfortably on 8–16 GB GPUs at Q4_K_M.
What quantization should I use for OLMo 2?
Q4_K_M is the best balance of quality and VRAM for OLMo 2 in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
How do I run OLMo 2 with Ollama?
Install Ollama, then run: ollama run olmo2:7b. This downloads OLMo 2 and starts a local, OpenAI-compatible endpoint — no internet connection is needed after the initial download.