---
type: "Article"
title: "Qwen3.8-Flash-Next 125B-A6B MLX 4/8-bit with mlx-serve on 1× Mac (M5 Ultra)"
description: "Qwen3.8-Flash-Next 125B-A6B MLX 4/8-bit with mlx-serve on 1× Mac (M5 Ultra): recipe by drowzeys with speed, condition and source for every number, weights and license."
resource: "https://www.contextstudios.ai/local-ai/drowzeys--mac-qwen38-flash-next-mlx-serve-mixed48"
language: "en"
generated:
  by: "process:contextstudios-md/1"
  at: "2026-10-24T16:13:50.705Z"
status: "stable"
---

# Qwen3.8-Flash-Next 125B-A6B MLX 4/8-bit with mlx-serve on 1× Mac (M5 Ultra)

Qwen3.8-Flash-Next 125B-A6B MLX 4/8-bit with mlx-serve on 1× Mac (M5 Ultra): 155 tok/s according to github.com (dataset as of Sep 29, 2026).

- Creator: [drowzeys](https://www.contextstudios.ai/local-ai/creators/drowzeys.md)
- Engine: mlx-serve 26.9.7-dev @933566c, MLX 0.32.3 (NAX an)
- Quantization: MLX mixed 4/8-bit (ddalcu, 100 GiB), KV bf16
- Model family: Qwen3.8-Flash-Next
- Hardware: mac x1
- Context: 262144
- Artificial Analysis Intelligence Index: 39.8 ([source](https://artificialanalysis.ai/models/qwen3-8-flash-next))

## All numbers

- 155 tok/s (Everyday; Everyday, 1 request, llmprobe standard prompt, thinking on, with MTP + PLD (per recipe)) — [source](https://github.com/drowzeys/keys-Independent-Bench-MLX-Serve-26.9.7-dev-Qwen3.8-Flash-Next-4bit-8bit-mixed#headline)
- decode_tps: 154.5 (1 request) — [source](https://github.com/drowzeys/keys-Independent-Bench-MLX-Serve-26.9.7-dev-Qwen3.8-Flash-Next-4bit-8bit-mixed#headline)
- decode_tps: 277.4 (1 request) — [source](https://github.com/drowzeys/keys-Independent-Bench-MLX-Serve-26.9.7-dev-Qwen3.8-Flash-Next-4bit-8bit-mixed#headline)
- prefill_tps: 4183 (1 request) — [source](https://github.com/drowzeys/keys-Independent-Bench-MLX-Serve-26.9.7-dev-Qwen3.8-Flash-Next-4bit-8bit-mixed#headline)
- ttft_ms: 174 (1 request) — [source](https://github.com/drowzeys/keys-Independent-Bench-MLX-Serve-26.9.7-dev-Qwen3.8-Flash-Next-4bit-8bit-mixed#headline)

## Weights

- [ddalcu/Qwen3.8-Flash-Next-MLX-Serve-mixed-4-8bit](https://huggingface.co/ddalcu/Qwen3.8-Flash-Next-MLX-Serve-mixed-4-8bit) (main, other)
- [Qwen/Qwen3.8-Flash-Next](https://huggingface.co/Qwen/Qwen3.8-Flash-Next) (base, other)

[https://github.com/drowzeys/keys-Independent-Bench-MLX-Serve-26.9.7-dev-Qwen3.8-Flash-Next-4bit-8bit-mixed](https://github.com/drowzeys/keys-Independent-Bench-MLX-Serve-26.9.7-dev-Qwen3.8-Flash-Next-4bit-8bit-mixed)

## Related

- [Local AI on your hardware.](https://www.contextstudios.ai/local-ai.md)