Qwen3.8-Flash-Next-MLX-4bit

Qwen3.8-Flash-Next MLX 4-bit

This is a reproducible MLX conversion of Qwen/Qwen3.8-Flash-Next at immutable revision f5d08274bafd880402bd16f5e3e6c514136ec06c.

MLX affine Q4/group-64 for eligible language, MTP, and vision matrices; the 160-wide n-gram table uses Q4/group-32. Routers, norms, biases, convolutions, and incompatible shapes remain dense.

• Artifact payload: 97.55 GiB
• Quantized Q2 modules: 0
• Quantized Q4/group-32 modules: 128
• Quantized Q4/group-64 modules: 927
• Source: 180B parameters including 125B main, 51B n-gram embedding, and 4B MTP
• Native context: 262,144 tokens

Runtime status

The tensor inventory, source hashes, MLX packing, fused-expert split, convolution layout, and zero-centered RMSNorm conversion are validated by the bundled MERERUN_CONVERSION.json. A Qwen4Exp-aware MLX runtime is required; do not expect an older mlx-lm or mlx-vlm release to dispatch this new architecture.

License

This redistribution retains the upstream Qwen Community License 1.0 in LICENSE. Review it before use. In particular, it contains attribution/display requirements for very large commercial products and separate-license conditions for certain commercial Model-as-a-Service and AI Work Assistant uses. The model is not gated; downloading or using it does not remove those terms.

The upstream model card is preserved as README.upstream.md.