community pool / model answer
Ornstein3.6-27B-MTP-NSC-ACE-SABER-GGUF
chat · GestaltLabs/Ornstein3.6-27B-MTP-NSC-ACE-SABER-GGUF
30.2 tok/s · model median · 17 runs in the model record.
| Reference rig | Quantization | Median result | Basis | TTFT (ms) | Peak VRAM | Max context | Engines |
|---|---|---|---|---|---|---|---|
| RTX 3090 24GB | 4-bit | 20.9 tok/s | measured | - | 22.6 GB | 262,144 | llama.cpp |
| RTX 3090 24GB ×2 | 4-bit | 49.5 tok/s | reported | 129 | - GB | 2,048 | llama.cpp |
| RTX 3090 24GB ×2 | 6-bit | 47.2 tok/s | measured | 202 | - GB | 2,048 | llama.cpp |
Context tested
The table above is the tested context for each exact rig and quantization cell. Empty fields remain empty rather than being filled with an estimate.