zeromodels/qwen3-next-80b-a3b-instruct

Run Qwen3-Next with Keras 3: JAX, PyTorch, or TensorFlow

kerasformers/qwen3-next-80b-a3b-instruct

Pure-Keras 3 conversion of Qwen/Qwen3-Next-80B-A3B-Instruct for kerasformers. One implementation runs unmodified on TensorFlow / Torch / JAX. Qwen3-Next is a Gated-DeltaNet + MoE hybrid; weights are stored in bfloat16.

For model details, license, and usage terms, see the upstream model card.

Paper: Qwen3 Technical Report (arXiv:2505.09388) · HF Papers

Paper: Qwen2.5-1M Technical Report (arXiv:2501.15383) · HF Papers

Paper: RULER: What's the Real Context Size of Your Long-Context Language Models? (arXiv:2404.06654) · HF Papers

Paper: YaRN: Efficient Context Window Extension of Large Language Models (arXiv:2309.00071) · HF Papers

✨ Quick start

import os
os.environ["KERAS_BACKEND"] = "torch"  # or "jax" / "tensorflow"

from kerasformers.models.qwen3_next import Qwen3NextTextGenerate, Qwen3NextTokenizer

model = Qwen3NextTextGenerate.from_weights("kerasformers/qwen3-next-80b-a3b-instruct")
tokenizer = Qwen3NextTokenizer.from_weights("kerasformers/qwen3-next-80b-a3b-instruct")

inputs = tokenizer("Give me a short introduction to large language models.")
outputs = model.generate(**inputs, max_new_tokens=128)
print(tokenizer.decode(outputs[0]))

Special Thanks

A huge thank you to the Qwen team at Alibaba for creating and releasing these models.

License: Apache 2.0.