Add official Gemma 4 12B-it plus QAT-INT4/INT8 catalog entries (with their GGUF sources), QAT quantization support across the quant tables and the prequantized-prefix list, and the missing RTX 3050 / 3050 Ti memory bandwidth so speed estimates stop falling back to the generic cuda value. |
||
|---|---|---|
| .. | ||
| docs | ||
| faces | ||
| hwfit | ||
| memory | ||
| research | ||
| search | ||
| shell | ||
| stt | ||
| tts | ||
| youtube | ||
| __init__.py | ||