NVIDIA RTX 3060 12GB vs AMD RX 7900 XT for local LLMs
46 reports on the NVIDIA RTX 3060 12GB and 5 on the AMD RX 7900 XT
Which is faster for local LLMs?
On Qwen2.5 7B at 4-bit, the NVIDIA RTX 3060 12GB runs at 49.0 tokens per second (one run) and the AMD RX 7900 XT at 115.0 (one run). Those are community reports whose engines and settings differ, so the gap is a rough sign, not a controlled measure of the two cards. Memory bandwidth, which caps how fast a card can write, is 2.2x higher on the AMD RX 7900 XT (360 GB/s on the NVIDIA RTX 3060 12GB, 800 on the AMD RX 7900 XT).
- Memory
- 12GB
- Memory bandwidth
- 360 GB/s
- FP16 compute
- 51 TFLOPS
- Reports
- 46
- Memory
- 20GB
- Memory bandwidth
- 800 GB/s
- FP16 compute
- 103 TFLOPS
- Reports
- 5
Measured on both
Median tokens per second from plain runs (one device, one request, no speculative decoding, the whole model in the device's memory) of the same model size at the same quant level. Engines and context lengths can still differ between the runs. How to read these
| Model | NVIDIA RTX 3060 12GB | AMD RX 7900 XT |
|---|---|---|
| Qwen2.5 7B4-bit | 49.0one run | 115.0one run |
Estimated by the calculator
The quant the calculator recommends for each card and its estimated speed at a 32,768-token context with 32 GB of system RAM, from memory bandwidth and the model's shape. Where a model was also measured above, the measurement wins.
| Model | NVIDIA RTX 3060 12GB | AMD RX 7900 XT |
|---|---|---|
| Qwen2.5 7B | 56.6 t/sQ4_K_S | 62.9 t/sQ8_0 |
| Qwen3.8 27B | 1.5 t/sQ8_0, layers on CPU | 37.2 t/sQ3_K_M |
| Qwen3.8 125B · 6B active | doesn't fit at these settings | 19.2 t/sQ2_K, experts in RAM |
| DeepSeek V4 Flash 284B · 13B active | doesn't fit at these settings | doesn't fit at these settings |
| Qwen3.6 35B · 3B active | 13.1 t/sQ6_K, experts in RAM | 147.9 t/sQ3_K_S |
| Qwen3.6 27B | 1.5 t/sQ8_0, layers on CPU | 37.2 t/sQ3_K_M |
| Gemma 4 26B · 4B active | 58.8 t/sQ2_K | 83.9 t/sQ4_K_S |
Change the context, RAM or model in the calculator: NVIDIA RTX 3060 12GB · AMD RX 7900 XT
Every report on each card, with its source: NVIDIA RTX 3060 12GB · AMD RX 7900 XT