Searching protocol for "hqq"
Compress LLMs with HQQ: Fast, no calibration.
Compress LLMs for faster inference.
Compress LLMs with HQQ
Compress LLMs for faster inference.
Compress LLMs to 4-bit without calibration.
Quantize LLMs without calibration data.
Compress LLMs without calibration data.
Compress LLMs without calibration data.
Quantize LLMs fast, no calibration needed.