---
title: Qwen3-30B-A3B-Thinking-2507-FP8
canonical_url: "https://www.modelscope.cn/models/unsloth/Qwen3-30B-A3B-Thinking-2507-FP8"
md_url: "https://www.modelscope.cn/models/unsloth/Qwen3-30B-A3B-Thinking-2507-FP8.md"
repository: unsloth/Qwen3-30B-A3B-Thinking-2507-FP8
last_updated: 2025-07-30
license: apache-2.0
model_type:
- qwen3_moe
architectures:
- Qwen3MoeForCausalLM
base_model:
- Qwen/Qwen3-30B-A3B-Thinking-2507-FP8
base_model_relation: quantized
parameters: 30.5B
tensor_type:
- F8_E4M3
- BF16
library_name:
- safetensors
- pytorch
frameworks:
- pytorch
inference_backends:
- "deploy_task text/emb"
- "lmdeploy_turbomind 0.9.1"
- "sglang 0.5.2"
- "vllm 0.9.2"
downloads: 386
stars: 0
tags:
- unsloth
---
# Qwen3-30B-A3B-Thinking-2507-FP8
> Qwen3-30B-A3B-Thinking-2507-FP8 - unsloth 在 ModelScope 开源的模型。[!NOTE] Includes Unsloth chat template fixes! For llama.cpp, use --jinja
unsloth/Qwen3-30B-A3B-Thinking-2507-FP8 是 ModelScope 魔搭社区上的 30.5B 参数机器学习模型，采用 apache-2.0 许可，基于 Qwen/Qwen3-30B-A3B-Thinking-2507-FP8 构建，可用 deploy_task text/emb、lmdeploy_turbomind 0.9.1、sglang 0.5.2 部署。
- **Repository**: unsloth/Qwen3-30B-A3B-Thinking-2507-FP8
- **License**: apache-2.0
- **Parameters**: 30.5B
- **Base model**: Qwen/Qwen3-30B-A3B-Thinking-2507-FP8
- **Inference backends**: deploy_task text/emb, lmdeploy_turbomind 0.9.1, sglang 0.5.2, vllm 0.9.2
- **Tags**: unsloth
- **Downloads**: 386
- **Stars**: 0
- **Last updated**: 2025-07-30
Source: https://www.modelscope.cn/models/unsloth/Qwen3-30B-A3B-Thinking-2507-FP8
---
> [!NOTE]
> Includes Unsloth **chat template fixes**!
For `llama.cpp`, use `--jinja`
>
