community pool / model answer
LFM2.5-230M
chat · LiquidAI/LFM2.5-230M
597.8 tok/s · model median · 9 runs in the model record.
| Reference rig | Quantization | Median result | Basis | TTFT (ms) | Peak VRAM | Max context | Engines |
|---|---|---|---|---|---|---|---|
| Radeon AI Pro R9700 32GB | 4-bit | 596.6 tok/s | measured | 76 | - GB | 4,096 | hipfire |
| RTX 3090 24GB | 4-bit | 144.4 tok/s | reported | 248 | 2.2 GB | 2,048 | llama.cpp |
| RX 5700 XT 8GB | 4-bit | 279.3 tok/s | reported | 67 | - GB | 4,096 | hipfire |
| RX 9070 XT 16GB | 4-bit | 528 tok/s | reported | 67 | - GB | 4,096 | hipfire |
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.