community-verified · measured pool data
Can the RTX 5090 32GB run gemma-4-12B-coder-fable5-composer2.5-MTP-NVFP4?
Yes — community-measured at 182.7 tok/s (2 runs at 4-bit).
Verified single-stream runs on the RTX 5090 32GB from the localmaxxing public pool. Every number on this page is measured — nothing here is estimated.
| Quantization | Median tok/s | TTFT (ms) | Peak VRAM | Max context tested | Engines |
|---|---|---|---|---|---|
| 4-bit | 182.7 | 33.1 | — | 2,048 | vllm |