What are the hardware requirements for Qwen3.8 Flash Next REAP320 OQ3e DWQ MTP Vision MLX?

Qwen3.8 Flash Next REAP320 OQ3e DWQ MTP Vision MLX requires approximately 70.6 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Qwen3.8 Flash Next REAP320 OQ3e DWQ MTP Vision MLX and how large is it?

Qwen3.8 Flash Next REAP320 OQ3e DWQ MTP Vision MLX is developed by Litwein, a model with 133.7b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Qwen3.8 Flash Next REAP320 OQ3e DWQ MTP Vision MLX?

Qwen3.8 Flash Next REAP320 OQ3e DWQ MTP Vision MLX is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Qwen3.8 Flash Next REAP320 OQ3e DWQ MTP Vision MLX by Litwein — VRAM 70.6GB

Name: Qwen3.8 Flash Next REAP320 OQ3e DWQ MTP Vision MLX
Author: Litwein

Qwen3.8 Flash Next REAP320 OQ3e DWQ MTP Vision MLX is an open-source language model by Litwein. Features: 133.7b LLM, VRAM: 70.6GB, License: other, Instruction-Based, LLM Explorer Score: 0.32.

3-bit Base model:quantized:qwen/qwen... Base model:qwen/qwen3.8-flash-... Conversational Dataset:agentsea/wave-ui-25k Dataset:allenai/tulu-3-sft-mix... Dataset:bigcode/self-oss-instr... Dataset:detection-datasets/coc... Dataset:huggingfacem4/chartqa Dataset:huggingfacem4/document... Dataset:lmms-lab/textvqa Dataset:nvidia/opencodereasoni... Dataset:open-r1/openr1-math-22... Dataset:open-r1/verifiable-cod... Dataset:open-thoughts/openthou... Dataset:swe-bench/swe-smith-tr... Dwq En Expert-pruning Image-text-to-text Instruct M4q Mixed-precision Mlx Moe Mtp Multimodal Omlx Oq Quantization Qwen3.8 Qwen4 exp Reap Region:us Ru Safetensors Sharded Speculative-decoding Tensorflow Vision

Model Card on HF 🤗: https://huggingface.co/Litwein/Qwen3.8-Flash-Next-REAP320-oQ3e-DWQ-MTP-Vision-MLX

Qwen3.8 Flash Next REAP320 OQ3e DWQ MTP Vision MLX Parameters and Internals

LLM Name	Qwen3.8 Flash Next REAP320 OQ3e DWQ MTP Vision MLX
Repository 🤗	https://huggingface.co/Litwein/Qwen3.8-Flash-Next-REAP320-oQ3e-DWQ-MTP-Vision-MLX
Base Model(s)	Qwen/Qwen3.8-Flash-Next Qwen/Qwen3.8-Flash-Next
Model Size	133.7b
Required VRAM	70.6 GB
Updated	2026-09-18
Maintainer	Litwein
Model Type	qwen4_exp
Instruction-Based	Yes
Model Files	5.2 GB: 1-of-14 5.2 GB: 2-of-14 5.2 GB: 3-of-14 5.2 GB: 4-of-14 5.2 GB: 5-of-14 5.2 GB: 6-of-14 5.2 GB: 7-of-14 5.2 GB: 8-of-14 5.3 GB: 9-of-14 5.3 GB: 10-of-14 5.3 GB: 11-of-14 5.3 GB: 12-of-14 5.3 GB: 13-of-14 2.5 GB: 14-of-14
Supported Languages	en ru
Model Architecture	Qwen4ExpForConditionalGeneration
License	other
Model Max Length	262144
Transformers Version	5.8.0.dev0
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Errors	replace

Rank the Qwen3.8 Flash Next REAP320 OQ3e DWQ MTP Vision MLX capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation