---
type: "Article"
title: "Qwen3.8-Flash-Next EXL3 3.05 bpw with Strata on 1× RTX 3090 (USB dock) + Mac mini M4"
description: "Qwen3.8-Flash-Next EXL3 3.05 bpw with Strata on 1× RTX 3090 (USB dock) + Mac mini M4: recipe by 0xSero with speed, condition and source for every number, weights and license."
resource: "https://www.contextstudios.ai/local-ai/0xsero--qwen38-flash-next-exl3-3.05bpw-strata-1x3090-usb-mac-mini-m4"
language: "en"
generated:
  by: "process:contextstudios-md/1"
  at: "2026-11-11T03:21:10.572Z"
status: "stable"
---

# Qwen3.8-Flash-Next EXL3 3.05 bpw with Strata on 1× RTX 3090 (USB dock) + Mac mini M4

Qwen3.8-Flash-Next EXL3 3.05 bpw with Strata on 1× RTX 3090 (USB dock) + Mac mini M4: 33.8 tok/s according to github.com (dataset as of Nov 3, 2026).

- Creator: [0xSero](https://www.contextstudios.ai/local-ai/creators/0xsero.md)
- Engine: Strata v0.1.41 (Niko1221, fb58e0dbc839) + Patches dieses Repos (Engine strata_o37u, Config LS) unter macOS über tgcu-CUDA-Shim auf tinygrads NV-Backend (3090 per USB); kalte Experten per Metal auf dem Mac mini M4 (16 GiB)
- Quantization: EXL3 3.05 bpw (turboderp, Branch 3.05bpw_h5_ng5; alle 512 Experten, nichts geprunt), Dense-Teil + mmproj IQ3_S GSQ-RCO GGUF, KV k8v4; Experten verteilt auf 3090-VRAM, Mac-mini-RAM (LRU 6.400 Slots) und NVMe
- Model family: Qwen3.8-Flash-Next
- Hardware: rtx-3090 x1
- Context: 204800
- Artificial Analysis Intelligence Index: 39.8 ([source](https://artificialanalysis.ai/models/qwen3-8-flash-next))

## All numbers

- 33.8 tok/s (Mixed; Mixed, 1 request, mean over 29 realistic document prompts (3.7–19.8K tokens), with MTP (per recipe)) — [source](https://github.com/sybil-solutions/qwen38-flash-next-3090-usb-offload/blob/main/README.md)
- decode_tps: 33.83 (1 request) — [source](https://github.com/sybil-solutions/qwen38-flash-next-3090-usb-offload/blob/main/README.md)
- prefill_tps: 1048.5 (1 request) — [source](https://github.com/sybil-solutions/qwen38-flash-next-3090-usb-offload/blob/main/README.md)

## Weights

- [turboderp/Qwen3.8-Flash-Next-exl3](https://huggingface.co/turboderp/Qwen3.8-Flash-Next-exl3/tree/3.05bpw_h5_ng5) (main, other)
- [ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF](https://huggingface.co/ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF) (other, apache-2.0)
- [Qwen/Qwen3.8-Flash-Next](https://huggingface.co/Qwen/Qwen3.8-Flash-Next) (mtp, other)

[https://github.com/sybil-solutions/qwen38-flash-next-3090-usb-offload](https://github.com/sybil-solutions/qwen38-flash-next-3090-usb-offload)

## Related

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