NVIDIA RTX 5090 vs AMD RX 7900 XTX for local LLMs
170 reports on the NVIDIA RTX 5090 and 47 on the AMD RX 7900 XTX
Which is faster for local LLMs?
On Qwen3.8 27B at 4-bit, the NVIDIA RTX 5090 has 80.0 generation tokens/s (7 runs); the AMD RX 7900 XTX has 42.5 generation tokens/s (6 runs); prompt processing 640.0 tokens/s (4 runs). 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 1.9x higher on the NVIDIA RTX 5090 (1,792 GB/s on the NVIDIA RTX 5090, 960 on the AMD RX 7900 XTX). Across 4 shared models with generation measurements on both cards, the NVIDIA RTX 5090's median is higher on 4 and the AMD RX 7900 XTX's on 0.
- Memory
- 32GB
- Memory bandwidth
- 1,792 GB/s
- FP16 compute
- 419 TFLOPS
- Reports
- 170
- Memory
- 24GB
- Memory bandwidth
- 960 GB/s
- FP16 compute
- 123 TFLOPS
- Reports
- 47
Measured on both
Generation and prompt processing speeds 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 bit class. Each side shows the engines and reported context behind its median. Context can mean a configured window or prompt depth, so these are community comparisons with differing settings. How to read these
| Model | NVIDIA RTX 5090 | AMD RX 7900 XTX |
|---|---|---|
| Qwen3.8 27B4-bit | 80.07 runsllama.cpp, TensorSharp +1, 2 not stated40K to 256K context, 2 not stated | 42.56 runsPP 640.0 (4 runs)llama.cpp, Ollama, 2 not stated64K to 220K context, 3 not stated |
| Gemma 4 31B4-bit | 64.0one runOllama32K context | 30.1one runPP 422.3 (one run)llama.cpp32K context |
| Gemma 4 26B · 4B active4-bit | 243.9one runPP 13809.2 (one run)llama.cppcontext not stated | 142.8one runPP 4331.2 (one run)llama.cppcontext not stated |
| Qwen3.6 35B · 3B active4-bit | 215.1one runllama.cppcontext not stated | 125.5one runPP 1874.7 (one run)llama.cppcontext not stated |
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 5090 | AMD RX 7900 XTX |
|---|---|---|
| Qwen3.8 27B | 44.9 t/sQ6_K | 34.8 t/sQ4_K_M |
| Gemma 4 31B | 46.2 t/sQ5_K_M | 34.1 t/sQ4_K_S |
| Gemma 4 26B · 4B active | 114.5 t/sQ6_K | 78.4 t/sQ5_K_S |
| Qwen3.6 35B · 3B active | 163.0 t/sQ5_K_M | 123.2 t/sQ4_K_S |
| Qwen3.8 125B · 6B active | 20.3 t/sQ2_K, experts in RAM | 19.5 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 27B | 44.9 t/sQ6_K | 34.8 t/sQ4_K_M |
| DeepSeek V4.1 Flash 552B · 16B active | doesn't fit at these settings | doesn't fit at these settings |
| GLM-5.3 320B · 18B active | doesn't fit at these settings | doesn't fit at these settings |
| Qwen3.5 9B | 54.8 t/sFP16 | 58.7 t/sQ8_0 |
Change the context, RAM or model in the calculator: NVIDIA RTX 5090 · AMD RX 7900 XTX
Every report on each card, with its source: NVIDIA RTX 5090 · AMD RX 7900 XTX