--- title: Qwen3-1.7B-w4a8-i8mm-kai canonical_url: "https://www.modelscope.cn/models/mllmTeam/Qwen3-1.7B-w4a8-i8mm-kai" md_url: "https://www.modelscope.cn/models/mllmTeam/Qwen3-1.7B-w4a8-i8mm-kai.md" repository: mllmTeam/Qwen3-1.7B-w4a8-i8mm-kai last_updated: 2025-10-30 architectures: - Qwen3ForCausalLM inference_backends: - "deploy_task text/emb" - "lmdeploy_turbomind 0.9.1" - "sglang 0.5.2" - "vllm 0.9.2" downloads: 97 stars: 0 --- # Qwen3-1.7B-w4a8-i8mm-kai > Qwen3-1.7B-w4a8-i8mm-kai - mllmTeam 在 ModelScope 开源的模型。Qwen3-1.7B-w4a8-i8mm-kai mllmTeam/Qwen3-1.7B-w4a8-i8mm-kai 是 ModelScope 魔搭社区上的机器学习模型，可用 deploy_task text/emb、lmdeploy_turbomind 0.9.1、sglang 0.5.2 部署。 - **Repository**: mllmTeam/Qwen3-1.7B-w4a8-i8mm-kai - **Inference backends**: deploy_task text/emb, lmdeploy_turbomind 0.9.1, sglang 0.5.2, vllm 0.9.2 - **Downloads**: 97 - **Stars**: 0 - **Last updated**: 2025-10-30 Source: https://www.modelscope.cn/models/mllmTeam/Qwen3-1.7B-w4a8-i8mm-kai --- Qwen3-1.7B-w4a8-i8mm-kai ```text ParameterFile[312 params, version=V2]: +---------------------------------------------------+------------------+-----------+ | name | shape | dtype | +===================================================+==================+===========+ | lm_head.weight | [151936, 2048] | Float32 | | lm_head_out.weight | [176245760] | UInt8 | | model.embed_tokens.weight | [151936, 2048] | Float32 | | model.layers.0.input_layernorm.weight | [2048] | Float32 | | model.layers.0.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.0.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.0.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.0.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.0.self_attn.k_norm.weight | [128] | Float32 | | model.layers.0.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.0.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.0.self_attn.q_norm.weight | [128] | Float32 | | model.layers.0.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.0.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.1.input_layernorm.weight | [2048] | Float32 | | model.layers.1.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.1.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.1.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.1.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.1.self_attn.k_norm.weight | [128] | Float32 | | model.layers.1.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.1.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.1.self_attn.q_norm.weight | [128] | Float32 | | model.layers.1.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.1.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.10.input_layernorm.weight | [2048] | Float32 | | model.layers.10.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.10.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.10.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.10.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.10.self_attn.k_norm.weight | [128] | Float32 | | model.layers.10.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.10.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.10.self_attn.q_norm.weight | [128] | Float32 | | model.layers.10.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.10.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.11.input_layernorm.weight | [2048] | Float32 | | model.layers.11.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.11.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.11.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.11.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.11.self_attn.k_norm.weight | [128] | Float32 | | model.layers.11.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.11.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.11.self_attn.q_norm.weight | [128] | Float32 | | model.layers.11.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.11.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.12.input_layernorm.weight | [2048] | Float32 | | model.layers.12.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.12.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.12.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.12.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.12.self_attn.k_norm.weight | [128] | Float32 | | model.layers.12.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.12.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.12.self_attn.q_norm.weight | [128] | Float32 | | model.layers.12.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.12.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.13.input_layernorm.weight | [2048] | Float32 | | model.layers.13.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.13.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.13.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.13.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.13.self_attn.k_norm.weight | [128] | Float32 | | model.layers.13.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.13.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.13.self_attn.q_norm.weight | [128] | Float32 | | model.layers.13.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.13.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.14.input_layernorm.weight | [2048] | Float32 | | model.layers.14.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.14.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.14.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.14.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.14.self_attn.k_norm.weight | [128] | Float32 | | model.layers.14.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.14.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.14.self_attn.q_norm.weight | [128] | Float32 | | model.layers.14.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.14.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.15.input_layernorm.weight | [2048] | Float32 | | model.layers.15.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.15.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.15.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.15.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.15.self_attn.k_norm.weight | [128] | Float32 | | model.layers.15.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.15.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.15.self_attn.q_norm.weight | [128] | Float32 | | model.layers.15.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.15.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.16.input_layernorm.weight | [2048] | Float32 | | model.layers.16.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.16.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.16.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.16.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.16.self_attn.k_norm.weight | [128] | Float32 | | model.layers.16.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.16.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.16.self_attn.q_norm.weight | [128] | Float32 | | model.layers.16.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.16.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.17.input_layernorm.weight | [2048] | Float32 | | model.layers.17.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.17.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.17.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.17.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.17.self_attn.k_norm.weight | [128] | Float32 | | model.layers.17.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.17.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.17.self_attn.q_norm.weight | [128] | Float32 | | model.layers.17.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.17.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.18.input_layernorm.weight | [2048] | Float32 | | model.layers.18.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.18.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.18.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.18.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.18.self_attn.k_norm.weight | [128] | Float32 | | model.layers.18.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.18.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.18.self_attn.q_norm.weight | [128] | Float32 | | model.layers.18.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.18.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.19.input_layernorm.weight | [2048] | Float32 | | model.layers.19.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.19.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.19.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.19.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.19.self_attn.k_norm.weight | [128] | Float32 | | model.layers.19.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.19.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.19.self_attn.q_norm.weight | [128] | Float32 | | model.layers.19.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.19.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.2.input_layernorm.weight | [2048] | Float32 | | model.layers.2.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.2.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.2.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.2.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.2.self_attn.k_norm.weight | [128] | Float32 | | model.layers.2.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.2.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.2.self_attn.q_norm.weight | [128] | Float32 | | model.layers.2.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.2.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.20.input_layernorm.weight | [2048] | Float32 | | model.layers.20.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.20.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.20.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.20.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.20.self_attn.k_norm.weight | [128] | Float32 | | model.layers.20.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.20.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.20.self_attn.q_norm.weight | [128] | Float32 | | model.layers.20.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.20.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.21.input_layernorm.weight | [2048] | Float32 | | model.layers.21.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.21.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.21.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.21.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.21.self_attn.k_norm.weight | [128] | Float32 | | model.layers.21.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.21.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.21.self_attn.q_norm.weight | [128] | Float32 | | model.layers.21.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.21.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.22.input_layernorm.weight | [2048] | Float32 | | model.layers.22.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.22.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.22.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.22.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.22.self_attn.k_norm.weight | [128] | Float32 | | model.layers.22.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.22.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.22.self_attn.q_norm.weight | [128] | Float32 | | model.layers.22.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.22.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.23.input_layernorm.weight | [2048] | Float32 | | model.layers.23.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.23.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.23.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.23.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.23.self_attn.k_norm.weight | [128] | Float32 | | model.layers.23.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.23.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.23.self_attn.q_norm.weight | [128] | Float32 | | model.layers.23.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.23.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.24.input_layernorm.weight | [2048] | Float32 | | model.layers.24.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.24.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.24.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.24.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.24.self_attn.k_norm.weight | [128] | Float32 | | model.layers.24.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.24.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.24.self_attn.q_norm.weight | [128] | Float32 | | model.layers.24.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.24.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.25.input_layernorm.weight | [2048] | Float32 | | model.layers.25.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.25.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.25.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.25.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.25.self_attn.k_norm.weight | [128] | Float32 | | model.layers.25.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.25.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.25.self_attn.q_norm.weight | [128] | Float32 | | model.layers.25.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.25.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.26.input_layernorm.weight | [2048] | Float32 | | model.layers.26.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.26.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.26.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.26.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.26.self_attn.k_norm.weight | [128] | Float32 | | model.layers.26.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.26.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.26.self_attn.q_norm.weight | [128] | Float32 | | model.layers.26.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.26.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.27.input_layernorm.weight | [2048] | Float32 | | model.layers.27.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.27.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.27.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.27.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.27.self_attn.k_norm.weight | [128] | Float32 | | model.layers.27.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.27.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.27.self_attn.q_norm.weight | [128] | Float32 | | model.layers.27.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.27.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.3.input_layernorm.weight | [2048] | Float32 | | model.layers.3.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.3.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.3.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.3.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.3.self_attn.k_norm.weight | [128] | Float32 | | model.layers.3.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.3.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.3.self_attn.q_norm.weight | [128] | Float32 | | model.layers.3.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.3.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.4.input_layernorm.weight | [2048] | Float32 | | model.layers.4.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.4.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.4.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.4.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.4.self_attn.k_norm.weight | [128] | Float32 | | model.layers.4.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.4.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.4.self_attn.q_norm.weight | [128] | Float32 | | model.layers.4.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.4.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.5.input_layernorm.weight | [2048] | Float32 | | model.layers.5.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.5.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.5.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.5.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.5.self_attn.k_norm.weight | [128] | Float32 | | model.layers.5.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.5.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.5.self_attn.q_norm.weight | [128] | Float32 | | model.layers.5.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.5.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.6.input_layernorm.weight | [2048] | Float32 | | model.layers.6.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.6.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.6.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.6.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.6.self_attn.k_norm.weight | [128] | Float32 | | model.layers.6.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.6.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.6.self_attn.q_norm.weight | [128] | Float32 | | model.layers.6.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.6.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.7.input_layernorm.weight | [2048] | Float32 | | model.layers.7.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.7.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.7.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.7.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.7.self_attn.k_norm.weight | [128] | Float32 | | model.layers.7.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.7.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.7.self_attn.q_norm.weight | [128] | Float32 | | model.layers.7.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.7.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.8.input_layernorm.weight | [2048] | Float32 | | model.layers.8.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.8.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.8.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.8.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.8.self_attn.k_norm.weight | [128] | Float32 | | model.layers.8.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.8.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.8.self_attn.q_norm.weight | [128] | Float32 | | model.layers.8.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.8.self_attn.v_proj.weight | [1187840] | UInt8 | | model.layers.9.input_layernorm.weight | [2048] | Float32 | | model.layers.9.mlp.down_proj.weight | [7094272] | UInt8 | | model.layers.9.mlp.gate_proj.weight | [7127040] | UInt8 | | model.layers.9.mlp.up_proj.weight | [7127040] | UInt8 | | model.layers.9.post_attention_layernorm.weight | [2048] | Float32 | | model.layers.9.self_attn.k_norm.weight | [128] | Float32 | | model.layers.9.self_attn.k_proj.weight | [1187840] | UInt8 | | model.layers.9.self_attn.o_proj.weight | [2375680] | UInt8 | | model.layers.9.self_attn.q_norm.weight | [128] | Float32 | | model.layers.9.self_attn.q_proj.weight | [2375680] | UInt8 | | model.layers.9.self_attn.v_proj.weight | [1187840] | UInt8 | | model.norm.weight | [2048] | Float32 | +---------------------------------------------------+------------------+-----------+ ```