# GeForce RTX 3060 (12 GB)

> OnPrem World · https://onpremworld.com/en/hardware/rtx-3060-12gb · Last updated: 2026-10-14 · CC BY 4.0

12 GB of VRAM for a $329 launch price: room for 8–14B models at 5- to 8-bit, though its 360 GB/s memory bandwidth keeps generation speed modest.

## Specifications

|  |  |
|---|---|
| Memory | 12 GB GDDR6 |
| Memory bandwidth | 360 GB/s |
| Power | 170 W |
| Launch | 2021-02 |
| Launch price | US$329 |
| Architecture | Ampere |
| Software | CUDA |

## What it can run (12 GB, 8K)

Estimates, not benchmarks. See the methodology.

| Model | Parameters | Best quant | Total | tok/s (est.) |
|---|---|---|---|---|
| Phi-4 (14B) | 14B | Q5_K_M | 11.8 GiB | 25 |
| Gemma 3 12B IT | 12B | Q6_K | 11.0 GiB | 25 |
| Gemma 4 12B Unified IT | 11.95B | Q6_K | 10.5 GiB | 26 |
| Qwen3.5-9B | 9B | Q8_0 | 10.1 GiB | 26 |
| Qwen3-8B | 8.2B | Q8_0 | 10.1 GiB | 29 |
| DeepSeek-R1-0528-Qwen3-8B | 8.19B | Q8_0 | 10.1 GiB | 29 |
| Llama 3.1 8B Instruct | 8B | Q8_0 | 9.8 GiB | 30 |
| Gemma 4 E4B IT | 8B | Q8_0 | 9.1 GiB | 30 |
| Llama 3.2 3B Instruct | 3.21B | FP16 | 7.7 GiB | 39 |
| Kimi K2.6 | 1000B | Doesn't fit | – | – |
| DeepSeek-R1-0528 (671B) | 671B | Doesn't fit | – | – |
| DeepSeek-V4-Flash | 284B | Doesn't fit | – | – |
| Qwen3.8-Flash-Next | 180B | Doesn't fit | – | – |
| Qwen3.5-122B-A10B | 122B | Doesn't fit | – | – |
| Mistral Small 4 119B (2603) | 119B | Doesn't fit | – | – |
| gpt-oss-120b | 117B | Doesn't fit | – | – |
| Llama 4 Scout 17B-16E Instruct | 109B | Doesn't fit | – | – |
| GLM-4.5-Air | 106B | Doesn't fit | – | – |
| Qwen3-Coder-Next (80B-A3B) | 80B | Doesn't fit | – | – |
| Llama 3.3 70B Instruct | 70B | Doesn't fit | – | – |
| Qwen3.5-35B-A3B | 35B | Doesn't fit | – | – |
| Qwen3.6-35B-A3B | 35B | Doesn't fit | – | – |
| Qwen3-32B | 32.8B | Doesn't fit | – | – |
| DeepSeek-R1-Distill-Qwen-32B | 32.76B | Doesn't fit | – | – |
| Gemma 4 31B IT | 30.7B | Doesn't fit | – | – |
| Qwen3-30B-A3B (2507) | 30.5B | Doesn't fit | – | – |
| Qwen3-Coder-30B-A3B-Instruct | 30.5B | Doesn't fit | – | – |
| GLM-4.7-Flash (30B-A3B) | 30B | Doesn't fit | – | – |
| Qwen3.5-27B | 27B | Doesn't fit | – | – |
| Qwen3.6-27B | 27B | Doesn't fit | – | – |
| Qwen3.8-27B | 27B | Doesn't fit | – | – |
| Gemma 3 27B IT | 27B | Doesn't fit | – | – |
| Gemma 4 26B A4B IT | 25.2B | Doesn't fit | – | – |
| Mistral Small 3.2 24B Instruct (2506) | 24B | Doesn't fit | – | – |
| Magistral Small 1.2 (2509) | 24B | Doesn't fit | – | – |
| Devstral Small 2 24B Instruct (2512) | 24B | Doesn't fit | – | – |
| gpt-oss-20b | 21B | Doesn't fit | – | – |

## Prices

Latest price at each shop, as listed on the day we checked.

| Region | Shop | Listing | Price | Per GB | Checked |
|---|---|---|---|---|---|
| Taiwan | momo購物網 | 【MSI 微星】GeForce RTX 3060 VENTUS 2X 12G OC 顯示卡 | [NT$15,590](https://www.momoshop.com.tw/product/8528368) | NT$1,299/GB | 2026-10-14 |
| Taiwan | 原價屋 CoolPC | ｛微星 RTX3060 VENTUS 2X 12G OC｝1807MHz/23.5cm/雙風扇/註四年 (Line on the shop's public price list (no per-product page); list stamped 2026/10/14 15:47) | [NT$15,590](https://www.coolpc.com.tw/evaluate.php) | NT$1,299/GB | 2026-10-14 |
| US | Newegg | GIGABYTE WINDFORCE OC GeForce RTX 3060 12GB GDDR6 PCI Express 4.0 x16 ATX Graphics Card GV-N3060WF2OC-12GD REV2.0 | [US$459.99](https://www.newegg.com/gigabyte-windforce-oc-gv-n3060wf2oc-12gd-rev2-0-geforce-rtx-3060-12gb-graphics-card-double-fans/p/N82E16814932658) | US$38/GB | 2026-10-14 |

Taiwan value: about 52.0 tok/s on Llama 3.1 8B Instruct (Q4_K_M, 8K), or 3.3 tok/s per NT$1,000.

US value: about 52.0 tok/s on Llama 3.1 8B Instruct (Q4_K_M, 8K), or 11.3 tok/s per US$100.

All prices: https://onpremworld.com/en/prices.md

## Frequently asked questions

### Which LLMs can the GeForce RTX 3060 (12 GB) run?

With 12 GB and an 8K context, the largest models that fit include: Phi-4 (14B), Gemma 3 12B IT, Gemma 4 12B Unified IT, Qwen3.5-9B, Qwen3-8B. See the full table below.

### How fast is the GeForce RTX 3060 (12 GB) for local LLMs?

Generation speed is mostly set by memory bandwidth (360 GB/s). For Llama 3.1 8B Instruct at Q8_0, expect roughly 30 tokens/s; bigger models are slower.

### Can the GeForce RTX 3060 (12 GB) run 70B models?

Llama 3.3 70B Instruct does not fit on the GeForce RTX 3060 (12 GB) (12 GB), even at Q4_K_M with an 8K context.

## Sources

- https://www.nvidia.com/en-us/geforce/graphics-cards/30-series/rtx-3060-3060ti/
- https://www.nvidia.com/en-gb/geforce/news/game-on-you-asked-we-answered-qa/
- https://www.nvidia.com/en-us/geforce/news/geforce-rtx-3060-out-now/
- https://nvidianews.nvidia.com/news/nvidia-introduces-geforce-rtx-3060-next-generation-of-the-worlds-most-popular-gpu
- https://en.wikipedia.org/wiki/List_of_Nvidia_graphics_processing_units