AI가 움직이고 미래가 업데이트됩니다.Qwen3.6-35B-A3B-GPTQ-4bit
GPTQ 4-bit quantization of Qwen/Qwen3.6-35B-A3B, a 35B-parameter Mixture-of-Experts (MoE) multimodal model with 3B activated parameters per token. Includes full vision encoder and MTP (Multi-Token Prediction) module for image understanding and speculative decoding support. All 30,720 MoE expert modules (256 experts x 3 projections x 40 layers) are quantized to INT4 using GPTQ. Non-expert modules (including the full vision encoder and MTP module) remain at BF16/FP16 for quality preservation.
번역 준비 중입니다. 원문을 표시합니다.
GPTQ 4-bit quantization of Qwen/Qwen3.6-35B-A3B, a 35B-parameter Mixture-of-Experts (MoE) multimodal model with 3B activated parameters per token. Includes full vision encoder and MTP (Multi-Token Prediction) module for image understanding and speculative decoding support. All 30,720 MoE expert modules (256 experts x 3 projections x 40 layers) are quantized to INT4 using GPTQ. Non-expert modules (including the full vision encoder and MTP module) remain at BF16/FP16 for quality preservation.