What are the hardware requirements for DeepSeek V4 Flash Vision Exp Mlx OQ8e DSpark?

DeepSeek V4 Flash Vision Exp Mlx OQ8e DSpark requires approximately 166.1 GB of VRAM and supports a context window of 1024K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed DeepSeek V4 Flash Vision Exp Mlx OQ8e DSpark and how large is it?

DeepSeek V4 Flash Vision Exp Mlx OQ8e DSpark is developed by Solstice-AI, a model with 304.6b 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 DeepSeek V4 Flash Vision Exp Mlx OQ8e DSpark?

DeepSeek V4 Flash Vision Exp Mlx OQ8e DSpark is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

DeepSeek V4 Flash Vision Exp Mlx OQ8e DSpark by Solstice-AI — VRAM 166.1GB, 1024K context

Name: DeepSeek V4 Flash Vision Exp Mlx OQ8e DSpark
Author: Solstice-AI

DeepSeek V4 Flash Vision Exp Mlx OQ8e DSpark is an open-source language model by Solstice-AI. Features: 304.6b LLM, VRAM: 166.1GB, Context: 1024K, License: mit, LLM Explorer Score: 0.31.

8-bit Apple-silicon Base model:quantized:unsloth/d... Base model:unsloth/deepseek-v4... Conversational Deepseek Deepseek-v4 Deepseek v4 Dspark En Image-text-to-text Long-context Metal Mixed-precision Mlx Mtp Multimodal Omlx Oq8e Region:us Safetensors Sharded Solstice-ai Speculative-decoding Tensorflow Vision Zh

Model Card on HF 🤗: https://huggingface.co/Solstice-AI/DeepSeek-V4-Flash-Vision-Exp-mlx-oQ8e-DSpark

DeepSeek V4 Flash Vision Exp Mlx OQ8e DSpark Parameters and Internals

LLM Name	DeepSeek V4 Flash Vision Exp Mlx OQ8e DSpark
Repository 🤗	https://huggingface.co/Solstice-AI/DeepSeek-V4-Flash-Vision-Exp-mlx-oQ8e-DSpark
Base Model(s)	DeepSeek V4 Flash Vision Exp unsloth/DeepSeek-V4-Flash-Vision-Exp
Model Size	304.6b
Required VRAM	166.1 GB
Updated	2026-09-10
Maintainer	Solstice-AI
Model Type	deepseek_v4
Model Files	5.0 GB: 1-of-30 6.1 GB: 2-of-30 5.7 GB: 3-of-30 5.7 GB: 4-of-30 5.7 GB: 5-of-30 5.7 GB: 6-of-30 5.7 GB: 7-of-30 5.7 GB: 8-of-30 5.7 GB: 9-of-30 5.7 GB: 10-of-30 5.7 GB: 11-of-30 5.7 GB: 12-of-30 5.7 GB: 13-of-30 5.7 GB: 14-of-30 5.7 GB: 15-of-30 5.7 GB: 16-of-30 5.7 GB: 17-of-30 5.7 GB: 18-of-30 5.7 GB: 19-of-30 5.7 GB: 20-of-30 5.7 GB: 21-of-30 5.7 GB: 22-of-30 5.7 GB: 23-of-30 5.7 GB: 24-of-30 5.7 GB: 25-of-30 5.7 GB: 26-of-30 5.7 GB: 27-of-30 5.7 GB: 28-of-30 5.7 GB: 29-of-30 1.1 GB: 30-of-30
Supported Languages	en zh
Model Architecture	DeepseekV4ForCausalLM
License	mit
Context Length	1048576
Model Max Length	1048576
Transformers Version	5.16.0.dev0
Tokenizer Class	TokenizersBackend
Padding Token	<｜▁pad▁｜>
Vocabulary Size	129280
Torch Data Type	bfloat16

Rank the DeepSeek V4 Flash Vision Exp Mlx OQ8e DSpark 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

Best Alternatives to DeepSeek V4 Flash Vision Exp Mlx OQ8e DSpark

Best Alternatives	Context / RAM	Downloads	Likes
DeepSeek V4 Flash Vision Exp	1024K / 146 GB	17893	441
...Seek V4 Flash Vision Exp NVFP4	1024K / 116 GB	1601	1
... Flash Vision Exp MXFP4 DSpark	1024K / 142.4 GB	1051	1
DeepSeek V4 Flash Vision Exp	1024K / 146 GB	487	8
...ash Vision Exp Mlx OQ4e DSpark	1024K / 166 GB	323	1
...Seek V4 Flash Vision Exp NVFP4	1024K / 119.8 GB	146	1
...eek V4 Flash 0731 Vision NVFP4	1024K / 111.3 GB	238	15
...h Vision Exp Abliterated NVFP4	1024K / 123.6 GB	72	1
...Seek V4 Flash 0731 Vision BF16	1024K / 116.3 GB	78	3

Note: green Score (e.g. "73.2") means that the model is better than Solstice-AI/DeepSeek-V4-Flash-Vision-Exp-mlx-oQ8e-DSpark.