Qwen/Qwen3.6-35B-A3B config.json

Fetched 2026-10-18 from the same commit as the card (995ad96eacd98c81ed38be0c5b274b04031597b0). Top-level model_type qwen3_5_moe with a nested text_config of model_type qwen3_5_moe_text plus a vision_config. text_config: hidden_size 2048, 40 layers, 16 query heads, 2 kv heads, head_dim 256, 256 experts, 8 per token, moe_intermediate_size 512, shared_expert_intermediate_size 512, silu, vocab 248320, tie_word_embeddings false, max_position_embeddings 262144, full_attention_interval 4, layer_types 30 linear_attention + 10 full_attention, linear_conv_kernel_dim 4, linear_num_key_heads 16, linear_num_value_heads 32, linear key and value head dim 128. rope_parameters: rope_type default, rope_theta 10000000, partial_rotary_factor 0.25, mrope_interleaved true, mrope_section [11, 11, 10]. No rope_scaling key.