AtomicChat/Qwen3.5-9B-MLX-4bit

Base model: Qwen/Qwen3.5-9B

Qwen3.5 9B, self-quantized to MLX by Atomic Chat. Built straight from Qwen's original weights with a per-tensor importance matrix, so this is not a repack of somebody else's files. Runs fully offline.

Highlights

• 9.7B parameters: the weights this repo quantizes.
• Context length: 262,144 tokens (256K), as published by Qwen.
• 32 layers: Dense decoder.
• Modalities: Text, Image.
• Full imatrix ladder: every quant is calibrated with an importance matrix.
• Unified Vision-Language Foundation: Early fusion training on multimodal tokens achieves cross-generational parity with Qwen3 and outperforms Qwen3-VL models across reasoning, coding, agents, and visual understanding benchmarks.
• Efficient Hybrid Architecture: Gated Delta Networks combined with sparse Mixture-of-Experts deliver high-throughput inference with minimal latency and cost overhead.

These MLXs are self-quantized from the original weights, not a repack. The importance matrix keeps low-bit quants closer to the full-precision model.

Model Overview

Property	Value
Base model	Qwen/Qwen3.5-9B
Parameters	9.7B
Layers	32
Context length	262,144 tokens (256K)
Vocabulary	248,320
Modalities	Text, Image
Architecture	Dense decoder, 16 attention heads over 4 KV heads, Qwen3_5ForConditionalGeneration
This repo	MLX weights

Get started

• Atomic Chat: search AtomicChat/Qwen3.5-9B-MLX-4bit and hit Use this model.
• mlx-lm: mlx_lm.generate --model AtomicChat/Qwen3.5-9B-MLX-4bit --prompt "Hello" --max-tokens 512
• Server: mlx_lm.server --model AtomicChat/Qwen3.5-9B-MLX-4bit --port 8080

Best practices

Parameter	Value
temperature	1.0
top_p	0.95
top_k	20
min_p	0.0
repetition_penalty	1.0

Qwen's recommended sampling configuration for Qwen/Qwen3.5-9B.

How these were made

1. Download Qwen/Qwen3.5-9B (original weights).
2. Convert and quantize with mlx_lm.convert on our pipeline.

License

Original model by Qwen, released under the Apache 2.0 license. Full terms: Apache 2.0. Quantized by Atomic Chat.