---
title: Kimi-K2-Instruct-GPU-weight
canonical_url: "https://www.modelscope.cn/models/ApproachingAI2024/Kimi-K2-Instruct-GPU-weight"
md_url: "https://www.modelscope.cn/models/ApproachingAI2024/Kimi-K2-Instruct-GPU-weight.md"
repository: ApproachingAI2024/Kimi-K2-Instruct-GPU-weight
chinese_name: Kimi-K2-Instruct-GPU-weight
last_updated: 2025-08-13
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - deepseek_v3
architectures:
  - DeepseekV3ForCausalLM
parameters: 146.5B
tensor_type:
  - I64
  - I32
  - BF16
  - F8_E4M3
  - F32
library_name:
  - safetensors
  - pytorch
frameworks:
  - Pytorch
inference_backends:
  - "deploy_task text/emb"
  - "sglang 0.5.2"
  - "vllm 0.9.2"
downloads: 17
stars: 0
---

# Kimi-K2-Instruct-GPU-weight

> Kimi-K2-Instruct-GPU-weight - ApproachingAI2024 在 ModelScope 开源的模型。当前模型的精度为非 experts 部分FP8，experts 部分为 GPTQ4

ApproachingAI2024/Kimi-K2-Instruct-GPU-weight 是 ModelScope 魔搭社区上的 146.5B 参数text-generation模型，可用 deploy_task text/emb、sglang 0.5.2、vllm 0.9.2 部署。

- **Repository**: ApproachingAI2024/Kimi-K2-Instruct-GPU-weight
- **Tasks**: text-generation
- **Parameters**: 146.5B
- **Inference backends**: deploy_task text/emb, sglang 0.5.2, vllm 0.9.2
- **Downloads**: 17
- **Stars**: 0
- **Last updated**: 2025-08-13

Source: https://www.modelscope.cn/models/ApproachingAI2024/Kimi-K2-Instruct-GPU-weight

---

### 当前模型的精度为非 experts 部分FP8，experts 部分为 GPTQ4