NVIDIA RTX 5090 vs NVIDIA RTX Pro 4000 Blackwell SFF for local LLMs
154 reports on the NVIDIA RTX 5090 and 3 on the NVIDIA RTX Pro 4000 Blackwell SFF
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 4.1x higher on the NVIDIA RTX 5090 (1,792 GB/s on the NVIDIA RTX 5090, 432 on the NVIDIA RTX Pro 4000 Blackwell SFF). The estimates below compare them on the models people run most.
- Memory
- 32GB
- Memory bandwidth
- 1,792 GB/s
- FP16 compute
- 419 TFLOPS
- Reports
- 154
- Memory
- 24GB
- Memory bandwidth
- 432 GB/s
- FP16 compute
- 96.3 TFLOPS
- Reports
- 3
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 5090 | NVIDIA RTX Pro 4000 Blackwell SFF |
|---|---|---|
| Qwen3.8 27B | 44.9 t/sQ6_K | 15.6 t/sQ4_K_M |
| Qwen3.8 125B · 6B active | 20.3 t/sQ2_K, experts in RAM | 17.6 t/sQ2_K, experts in RAM |
| Qwen3.6 35B · 3B active | 163.0 t/sQ5_K_M | 55.4 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 | 44.9 t/sQ6_K | 15.6 t/sQ4_K_M |
| Gemma 4 26B · 4B active | 114.5 t/sQ6_K | 35.3 t/sQ5_K_S |
Change the context, RAM or model in the calculator: NVIDIA RTX 5090 · NVIDIA RTX Pro 4000 Blackwell SFF
Every report on each card, with its source: NVIDIA RTX 5090 · NVIDIA RTX Pro 4000 Blackwell SFF