# Can I run gpt-oss 120B on Nvidia GeForce RTX 3060 (12GB)?

> **No, not enough memory.** Needs ~62.4 GB even at Q4_K_M, but only ~11 GB is usable.

Last validated: 2026-06-15.

## Memory math
- gpt-oss 120B: 117B (MoE, 5.1B active); Q4_K_M weighs 59.03 GB on disk.
- At Q4_K_M with a 4k context it needs ~62.4 GB (weights + KV cache + ~0.8 GB runtime overhead, ±15% with context length).
- Nvidia GeForce RTX 3060 (12GB): 12 GB vram, ~11 GB usable for model weights.
- Headroom: ~-51.4 GB.

## How to run
Recommended quant: n/a. Best tool on Windows: LM Studio.

```sh
ollama run gpt-oss:120b
```

## Run gpt-oss 120B on other devices
- [Apple M4 Max (128GB)](https://localmodel.run/can-i-run/gpt-oss-120b/apple-m4-max-128gb)
- [Apple M5 Max (128GB)](https://localmodel.run/can-i-run/gpt-oss-120b/apple-m5-max-128gb)
- [Apple M3 Ultra (256GB)](https://localmodel.run/can-i-run/gpt-oss-120b/apple-m3-ultra-256gb)

## Other models that run on Nvidia GeForce RTX 3060 (12GB)
- [Phi-4 14B (14B)](https://localmodel.run/can-i-run/phi-4-14b/nvidia-rtx-3060-12gb)
- [Qwen2.5 14B (14B)](https://localmodel.run/can-i-run/qwen2.5-14b/nvidia-rtx-3060-12gb)
- [Qwen3 14B (14B)](https://localmodel.run/can-i-run/qwen3-14b/nvidia-rtx-3060-12gb)
- [DeepSeek-R1-Distill-Qwen 14B (14B)](https://localmodel.run/can-i-run/deepseek-r1-distill-qwen-14b/nvidia-rtx-3060-12gb)
- [Qwen2.5 Coder 14B (14B)](https://localmodel.run/can-i-run/qwen2.5-coder-14b/nvidia-rtx-3060-12gb)
- [Mistral Nemo 12B (12.2B)](https://localmodel.run/can-i-run/mistral-nemo-12b/nvidia-rtx-3060-12gb)

Estimate, not a guarantee. Sources: https://ollama.com/library/gpt-oss, https://ollama.com/library/gpt-oss/tags, https://huggingface.co/ggml-org/gpt-oss-120b-GGUF, https://huggingface.co/openai/gpt-oss-120b.
More: https://localmodel.run/can-i-run/gpt-oss-120b/nvidia-rtx-3060-12gb