{ "producer": { "name": "modelopt", "version": "0.43.0rc2.dev91+gc79ebc014" }, "quantization": { "quant_algo": "NVFP4", "kv_cache_quant_algo": "FP8", "group_size": 16, "exclude_modules": [ "lm_head", "model.embed_vision*", "model.language_model.layers.0.mlp*", "model.language_model.layers.0.router*", "model.language_model.layers.0.self_attn*", "model.language_model.layers.1.mlp*", "model.language_model.layers.1.router*", "model.language_model.layers.1.self_attn*", "model.language_model.layers.10.mlp*", "model.language_model.layers.10.router*", "model.language_model.layers.10.self_attn*", "model.language_model.layers.11.mlp*", "model.language_model.layers.11.router*", "model.language_model.layers.11.self_attn*", "model.language_model.layers.12.mlp*", "model.language_model.layers.12.router*", "model.language_model.layers.12.self_attn*", "model.language_model.layers.13.mlp*", "model.language_model.layers.13.router*", "model.language_model.layers.13.self_attn*", "model.language_model.layers.14.mlp*", "model.language_model.layers.14.router*", "model.language_model.layers.14.self_attn*", "model.language_model.layers.15.mlp*", "model.language_model.layers.15.router*", "model.language_model.layers.15.self_attn*", "model.language_model.layers.16.mlp*", "model.language_model.layers.16.router*", "model.language_model.layers.16.self_attn*", "model.language_model.layers.17.mlp*", "model.language_model.layers.17.router*", "model.language_model.layers.17.self_attn*", "model.language_model.layers.18.mlp*", "model.language_model.layers.18.router*", "model.language_model.layers.18.self_attn*", "model.language_model.layers.19.mlp*", "model.language_model.layers.19.router*", "model.language_model.layers.19.self_attn*", "model.language_model.layers.2.mlp*", "model.language_model.layers.2.router*", "model.language_model.layers.2.self_attn*", "model.language_model.layers.20.mlp*", "model.language_model.layers.20.router*", "model.language_model.layers.20.self_attn*", "model.language_model.layers.21.mlp*", "model.language_model.layers.21.router*", "model.language_model.layers.21.self_attn*", "model.language_model.layers.22.mlp*", "model.language_model.layers.22.router*", "model.language_model.layers.22.self_attn*", "model.language_model.layers.23.mlp*", "model.language_model.layers.23.router*", "model.language_model.layers.23.self_attn*", "model.language_model.layers.24.mlp*", "model.language_model.layers.24.router*", "model.language_model.layers.24.self_attn*", "model.language_model.layers.25.mlp*", "model.language_model.layers.25.router*", "model.language_model.layers.25.self_attn*", "model.language_model.layers.26.mlp*", "model.language_model.layers.26.router*", "model.language_model.layers.26.self_attn*", "model.language_model.layers.27.mlp*", "model.language_model.layers.27.router*", "model.language_model.layers.27.self_attn*", "model.language_model.layers.28.mlp*", "model.language_model.layers.28.router*", "model.language_model.layers.28.self_attn*", "model.language_model.layers.29.mlp*", "model.language_model.layers.29.router*", "model.language_model.layers.29.self_attn*", "model.language_model.layers.3.mlp*", "model.language_model.layers.3.router*", "model.language_model.layers.3.self_attn*", "model.language_model.layers.4.mlp*", "model.language_model.layers.4.router*", "model.language_model.layers.4.self_attn*", "model.language_model.layers.5.mlp*", "model.language_model.layers.5.router*", "model.language_model.layers.5.self_attn*", "model.language_model.layers.6.mlp*", "model.language_model.layers.6.router*", "model.language_model.layers.6.self_attn*", "model.language_model.layers.7.mlp*", "model.language_model.layers.7.router*", "model.language_model.layers.7.self_attn*", "model.language_model.layers.8.mlp*", "model.language_model.layers.8.router*", "model.language_model.layers.8.self_attn*", "model.language_model.layers.9.mlp*", "model.language_model.layers.9.router*", "model.language_model.layers.9.self_attn*", "model.vision_tower*" ] } }