RTX 3080 10 GB vs RTX 3060 12 GB for local AI
The RTX 3060 12 GB has 2 GB more memory, and that decides it for local AI: it runs a better file (or runs at all) for 32 of 49 models, and 17 run at 8-bit or better against 12.
RTX 3080 10 GB
- VRAM
- 10 GB
- Memory
- GDDR6X
- Bandwidth
- 760 GB/s
- Architecture
- Ampere
- FP8 compute
- No
- Launch price
- $699
VS
RTX 3060 12 GB
- VRAM
- 12 GB
- Memory
- GDDR6
- Bandwidth
- 360 GB/s
- Architecture
- Ampere
- FP8 compute
- No
- Launch price
- $329
01Model by model
| Model | RTX 3080 10 GB | RTX 3060 12 GB | ||
|---|---|---|---|---|
| image models | ||||
| FLUX.1 dev | Runs Q4_K_S | Runs Q5_K_S | ||
| FLUX.1 schnell | Runs Q4_K_S | Runs Q5_K_S | ||
| FLUX.1 Kontext | Runs Q4_K_M | Runs Q5_K_M | ||
| FLUX.1 Krea | Runs Q4_K_M | Runs Q5_K_M | ||
| FLUX.1 Fill | Runs Q4_K_S | Runs Q5_K_S | ||
| FLUX.2 dev | Not practical Q2_K | Offload only Q4_K_M | ||
| FLUX.2 klein 9B | Runs Q5_K_M | Runs well FP8 | ||
| FLUX.2 klein 4B | Runs well 16-bit | Runs well 16-bit | ||
| Krea 2 | Tight Q3_K_M | Runs Q5_K_M | ||
| Qwen-Image | Tight Q2_K | Tight Q2_K | ||
| Qwen-Image-Edit | Offload only Q2_K | Tight Q2_K | ||
| Qwen-Image 2.1 | Runs well Q8_0 | Runs well Q8_0 | ||
| Z-Image Turbo | Runs well Q8_0 | Runs well Q8_0 | ||
| Z-Image | Runs well Q8_0 | Runs well Q8_0 | ||
| Ideogram 4 | Offload only Q4_1 | Offload only Q4_1 | ||
| Boogu-Image | Offload only Q5_1 | Runs Q5_1 | ||
| ERNIE-Image | Runs Q6_K | Runs well Q8_0 | ||
| HiDream-O1 | Offload only FP8 | Runs well FP8 | ||
| Mage-Flow | Runs well INT8 | Runs well 16-bit | ||
| Lumina 2.0 | Runs well 16-bit | Runs well 16-bit | ||
| HiDream-I1 Full | Tight Q2_K | Tight Q3_K_M | ||
| HiDream-I1 | Tight Q2_K | Tight Q3_K_M | ||
| SD 3.5 Large | Runs Q5_1 | Runs well Q8_0 | ||
| SD 3.5 Medium | Runs well 16-bit | Runs well 16-bit | ||
| Chroma1-HD | Runs Q6_K | Runs well FP8 | ||
| SDXL | Runs well 16-bit | Runs well 16-bit | ||
| Illustrious / Pony | Runs well 16-bit | Runs well 16-bit | ||
| SD 1.5 | Runs well 16-bit | Runs well 16-bit | ||
| HunyuanImage 2.1 | Offload only Q2_K | Tight Q2_K | ||
| video models | ||||
| Wan 2.1 14B | Offload only Q3_K_M | Tight Q3_K_M | ||
| Wan 2.1 1.3B | Runs well 16-bit | Runs well 16-bit | ||
| Wan 2.1 I2V 480P | Offload only Q4_K_M | Offload only Q3_K_M | ||
| Wan 2.1 I2V 720P | Offload only Q4_K_M | Offload only Q4_K_M | ||
| Wan VACE 14B | Offload only Q4_K_M | Offload only Q3_K_S | ||
| Wan 2.2 T2V | Tight Q2_K | Tight Q3_K_M | ||
| Wan 2.2 I2V | Tight Q2_K | Tight Q3_K_M | ||
| Wan 2.2 5B | Runs well Q8_0 | Runs well Q8_0 | ||
| Wan 2.2 Animate | Offload only Q2_K | Tight Q2_K | ||
| Wan Animate 2 | Offload only Q2_K | Tight Q2_K | ||
| Wan 2.2 S2V | Offload only Q4_K_M | Offload only Q4_K_M | ||
| SCAIL-2 | Offload only Q4_K_M | Offload only Q2_K | ||
| HunyuanVideo 13B | Offload only Q3_K_M | Tight Q3_K_M | ||
| LTX-Video 13B | Tight Q2_K | Tight Q3_K_M | ||
| HunyuanVideo 1.5 | Runs Q4_K_M | Runs Q6_K | ||
| LTX-2 | Offload only Q4_K_M | Offload only Q2_K | ||
| LTX-2.3 | Offload only Q4_K_M | Offload only Q2_K | ||
| LTX-2.5 | Offload only Q4_K_M | Offload only Q2_K | ||
| MiniMax H3 | Offload only Q4_K_M | Offload only Q4_K_M | ||
| MiniMax H3 Pruned | Offload only Q4_K_M | Offload only Q4_K_M | ||
Highlighted: the GPU that runs a better file for that model. Launch prices are the maker’s original list prices, not today’s street prices. Calculated from real file sizes; how the numbers work.
02Score
12run well on RTX 3080 10 GB
17run well on RTX 3060 12 GB
0better on RTX 3080 10 GB
32better on RTX 3060 12 GB