majentik/Qwen3-Coder-Next-MLX-2bit

Qwen3-Coder-Next - MLX 2-bit

2-bit weight-quantized MLX version of Qwen/Qwen3-Coder-Next, Qwen's 80B-A3B agentic coding MoE (512 experts, 10 active; hybrid Gated DeltaNet + Gated Attention; 256k context). Only ~3B parameters are active per token, so it runs far faster than its 80B total suggests. Converted with mlx_lm — the canonical MLX runtime for qwen3_next — and smoke-verified (chat + code probes) on Apple Silicon with this exact payload before publishing. See PROVENANCE.md.

Approximate model size: ~25 GB

Model Specifications

Property	Value
Base Model	Qwen/Qwen3-Coder-Next
Parameters	80 billion total (~3 billion active per token)
Architecture	MoE, hybrid Gated DeltaNet + Gated Attention (qwen3_next)
Modality	Text-only (code-focused)
Context Length	256k tokens
License	Apache 2.0
Weight Quantization	2-bit affine, group size 64 (~25 GB)
Framework	MLX (Apple Silicon), mlx-lm >= 0.31

Quickstart

from mlx_lm import load, generate

model, tokenizer = load("majentik/Qwen3-Coder-Next-MLX-2bit")

prompt = tokenizer.apply_chat_template(
    [{"role": "user", "content": "Write a Python function that merges two sorted lists."}],
    add_generation_prompt=True, tokenize=False,
)
print(generate(model, tokenizer, prompt=prompt, max_tokens=512))

Or from the command line:

mlx_lm.generate --model majentik/Qwen3-Coder-Next-MLX-2bit --prompt "Refactor this function ..."

Variants in this family

Variant	Approx size	Use case
2bit(https://huggingface.co/majentik/Qwen3-Coder-Next-MLX-2bit)	~25 GB	Smallest; quality floor
3bit	~34 GB	Low-RAM Macs
4bit	~43 GB	Balanced default
5bit	~53 GB	Higher fidelity
6bit	~62 GB	Near-8bit quality
8bit	~81 GB	Reference fidelity

Smoke verification covers load + short-form generation quality gates only; it is not a benchmark. For maximum fidelity use the largest variant that fits your unified memory (leave ~20% headroom for KV cache).