qwen3-30b-a3b-instruct-2507 (Keras 3, bf16)

Pure-Keras 3 conversion of Qwen/Qwen3-30B-A3B-Instruct-2507 for kerasformers. Apache 2.0.

This repo holds the converted model weights and tokenizer.json.

from kerasformers.models.qwen3_moe import Qwen3MoeGenerate, Qwen3MoeTokenizer
model = Qwen3MoeGenerate.from_weights("qwen3-30b-a3b-instruct-2507")
tokenizer = Qwen3MoeTokenizer.from_weights("qwen3-30b-a3b-instruct-2507")

Downloads last month: 14

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for kerasformers/qwen3-30b-a3b-instruct-2507

Base model

Qwen/Qwen3-30B-A3B-Instruct-2507

Finetuned

(82)

this model