NVIDIA RTX 4090 vs NVIDIA RTX A5000 for local LLMs
56 reports on the NVIDIA RTX 4090 and 1 on the NVIDIA RTX A5000
Which is faster for local LLMs?
Nobody has reported the same model at the same quant on both cards yet, so there's no measured head-to-head. Memory bandwidth, which caps how fast a card can write, is 31% higher on the NVIDIA RTX 4090 (1,008 GB/s on the NVIDIA RTX 4090, 768 on the NVIDIA RTX A5000). The estimates below compare them on the models people run most.
NVIDIA RTX 4090
- Memory
- 24GB
- Memory bandwidth
- 1,008 GB/s
- FP16 compute
- 330 TFLOPS
- Reports
- 56
NVIDIA RTX A5000
- Memory
- 24GB
- Memory bandwidth
- 768 GB/s
- FP16 compute
- 111.1 TFLOPS
- Reports
- 1
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.
| Model | NVIDIA RTX 4090 | NVIDIA RTX A5000 |
|---|---|---|
| Qwen3.8 27B | 36.5 t/sQ4_K_M | 27.8 t/sQ4_K_M |
| Qwen3.8 125B · 6B active | 19.6 t/sQ2_K, experts in RAM | 19.1 t/sQ2_K, experts in RAM |
| Qwen3.6 35B · 3B active | 129.3 t/sQ4_K_S | 98.5 t/sQ4_K_S |
| DeepSeek V4 Flash 284B · 13B active | doesn't fit at these settings | doesn't fit at these settings |
| Qwen3.6 27B | 36.5 t/sQ4_K_M | 27.8 t/sQ4_K_M |
| Gemma 4 26B · 4B active | 82.3 t/sQ5_K_S | 62.7 t/sQ5_K_S |
Change the context, RAM or model in the calculator: NVIDIA RTX 4090 · NVIDIA RTX A5000
Every report on each card, with its source: NVIDIA RTX 4090 · NVIDIA RTX A5000