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 rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
Radeon AI Pro R9700 32GB4-bit596.6 tok/smeasured76- GB4,096hipfire
RTX 3090 24GB4-bit144.4 tok/sreported2482.2 GB2,048llama.cpp
RX 5700 XT 8GB4-bit279.3 tok/sreported67- GB4,096hipfire
RX 9070 XT 16GB4-bit528 tok/sreported67- GB4,096hipfire

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.