amd/K2-Horizon-375B-A23B-MXFP4开源了新模型
根据官方来源,amd/K2-Horizon-375B-A23B-MXFP4开源了新模型。详细信息请以原始来源为准。
证据来源
查看来源摘录
AMD amd/K2-Horizon-375B-A23B-MXFP4 amd/K2-Horizon-375B-A23B-MXFP4 https://huggingface.co/amd/K2-Horizon-375B-A23B-MXFP4
查看来源摘录
amd/K2-Horizon-375B-A23B-MXFP4 · Hugging Face Hugging Face (https://huggingface.co/) Models (https://huggingface.co/models) Datasets (https://huggingface.co/datasets) Spaces (https://huggingface.co/spaces) Buckets new (https://huggingface.co/storage) Docs (https://huggingface.co/docs) Enterprise (https://huggingface.co/enterprise) Pricing (https://huggingface.co/pricing) Website Tasks (https://huggingface.co/tasks) HuggingChat (https://huggingface.co/chat) Collections (https://huggingface.co/collections) Languages (https://huggingface.co/languages) Organizations (https://huggingface.co/organizations) Community Blog (https://huggingface.co/blog) Posts (https://huggingface.co/posts) Daily Papers (https://huggingface.co/papers) Hardware (https://huggingface.co/hardware) Learn (https://huggingface.co/learn) Discord (https://huggingface.co/join/discord) Forum (https://discuss.huggingface.co/) GitHub (https://github.com/huggingface) Solutions Team & Enterprise (https://huggingface.co/enterprise) Hugging Face PRO (https://huggingface.co/pro) Enterprise Support (https://huggingface.co/support) Inference Providers (https://huggingface.co/inference/models) Inference Endpoints (https://huggingface.co/inference-endpoints) Storage Buckets (https://huggingface.co/storage) Log In (https://huggingface.co/login) Sign Up (https://huggingface.co/join) (https://huggingface.co/amd) amd (https://huggingface.co/amd) / K2-Horizon-375B-A23B-MXFP4 (https://huggingface.co/amd/K2-Horizon-375B-A23B-MXFP4) Like 0 Follow AMD 3.32k Safetensors (https://huggingface.co/models?library=safetensors) k2_horizon (https://huggingface.co/models?other=k2_horizon) custom_code (https://huggingface.co/models?other=custom_code) 8-bit precision (https://huggingface.co/models?other=8-bit) quark (https://huggingface.co/models?other=quark) License: apache-2.0 Model card (https://huggingface.co/amd/K2-Horizon-375B-A23B-MXFP4) Files Files and versions xet (https://huggingface.co/amd/K2-Horizon-375B-A23B-MXFP4/tree/main) Community (https://huggingface.co/amd/K2-Horizon-375B-A23B-MXFP4/discussions) Copy to bucket new K2-Horizon-375B-A23B-MXFP4 (https://huggingface.co/amd/K2-Horizon-375B-A23B-MXFP4#k2-horizon-375b-a23b-mxfp4) Model Overview (https://huggingface.co/amd/K2-Horizon-375B-A23B-MXFP4#model-overview) Model Quantization (https://huggingface.co/amd/K2-Horizon-375B-A23B-MXFP4#model-quantization) Quantization script (https://huggingface.co/amd/K2-Horizon-375B-A23B-MXFP4#quantization-script) Deployment (https://huggingface.co/amd/K2-Horizon-375B-A23B-MXFP4#deployment) Use with vLLM (https://huggingface.co/amd/K2-Horizon-375B-A23B-MXFP4#use-with-vllm) Evaluation (https://huggingface.co/amd/K2-Horizon-375B-A23B-MXFP4#evaluation) Accuracy (https://huggingface.co/amd/K2-Horizon-375B-A23B-MXFP4#accuracy) Reproduction (https://huggingface.co/amd/K2-Horizon-375B-A23B-MXFP4#reproduction) (https://huggingface.co/amd/K2-Horizon-375B-A23B-MXFP4#k2-horizon-375b-a23b-mxfp4) K2-Horizon-375B-A23B-MXFP4 (https://huggingface.co/amd/K2-Horizon-375B-A23B-MXFP4#model-overview) Model Overview Model Architecture: K2HorizonForCausalLM Input: Text Output: Text Supported Hardware Microarchitecture: AMD MI355 / MI350 (gfx950) ROCm: 7.2.3 PyTorch: 2.12.0 Transformers: 5.17.0 Operating System(s): Linux Inference Engine: vLLM (https://docs.vllm.ai/en/latest/) Model Optimizer: AMD-Quark (https://quark.docs.amd.com/latest/index.html) (v0.12.0) Weight quantization: routed MOE-only, OCP MXFP4, Static Activation quantization: routed MOE-only, OCP MXFP4, Dynamic (https://huggingface.co/amd/K2-Horizon-375B-A23B-MXFP4#model-quantization) Model Quantization Quantized from IFM/K2-Horizon-375B-A23B (https://huggingface.co/IFM/K2-Horizon-375B-A23B) with AMD Quark (https://quark.docs.amd.com/latest/index.html) . The routed MoE experts are quantized to OCP MXFP4, using static weight quantization and dynamic activation quantization. All remaining modules are k
来自 星盘大模型百科