---
title: DeepSeek-V4-Flash-MLX-Q2i
canonical_url: "https://www.modelscope.ai/models/inferencerlabs/DeepSeek-V4-Flash-MLX-Q2i"
md_url: "https://www.modelscope.ai/models/inferencerlabs/DeepSeek-V4-Flash-MLX-Q2i.md"
repository: inferencerlabs/DeepSeek-V4-Flash-MLX-Q2i
last_updated: 2026-09-02
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - deepseek_v4
architectures:
  - DeepseekV4ForCausalLM
base_model:
  - deepseek-ai/DeepSeek-V4-Flash
base_model_relation: quantized
parameters: 32.8B
tensor_type:
  - F32
  - U32
  - I64
  - BF16
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
downloads: 71
stars: 0
tags:
  - quantized
  - mlx
---

# DeepSeek-V4-Flash-MLX-Q2i

> DeepSeek-V4-Flash-MLX-Q2i - An open-source model by inferencerlabs on ModelScope. See DeepSeek-V4-Flash MLX in action - demonstration videos Tested with an M3 Ultra 512 GiB using Inferencer app v1.11.1 Text inference: 30 tokens/s @ 2000 tokens 95.49 GiB…

inferencerlabs/DeepSeek-V4-Flash-MLX-Q2i is a 32.8B-parameter text-generation model on ModelScope. derived from deepseek-ai/DeepSeek-V4-Flash.

- **Repository**: inferencerlabs/DeepSeek-V4-Flash-MLX-Q2i
- **Tasks**: text-generation
- **Parameters**: 32.8B
- **Base model**: deepseek-ai/DeepSeek-V4-Flash
- **Tags**: quantized, mlx
- **Downloads**: 71
- **Stars**: 0
- **Last updated**: 2026-09-02

Source: https://www.modelscope.ai/models/inferencerlabs/DeepSeek-V4-Flash-MLX-Q2i

---

**See DeepSeek-V4-Flash MLX in action - [demonstration videos](https://youtube.com/xcreate)**
   
#### Tested with an M3 Ultra 512 GiB using [Inferencer app v1.11.1](https://inferencer.com)
- Text inference: ~30 tokens/s @ 2000 tokens ~95.49 GiB (debug build)

<p style="margin-bottom:0px;">
<strong>Q2.8-INF uses the data-agnostic INF method tuned to yield maximum general accuracy within a 128 GiB memory budget</strong>

This build compresses the model to operate within the memory constraints of a 128 GiB RAM system while maintaining response coherence. However, overall accuracy is degraded due to the level of compression.
</p>

![Screenshot](https://cdn-uploads.huggingface.co/production/uploads/688479d616f1ec82fa645019/TphT4hr5CABuKO1MGccqh.jpeg)

##### Quantized with a modified version of [MLX](https://github.com/ml-explore/mlx)
##### For more details see our [demonstration videos](https://youtube.com/xcreate) or visit [DeepSeek-V4-Flash](https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro).

## Disclaimer

We are not the creator, originator, or owner of any model listed. Each model is created and provided by third parties. Models may not always be accurate or contextually appropriate. You are responsible for verifying the information before making important decisions. We are not liable for any damages, losses, or issues arising from its use, including data loss or inaccuracies in AI-generated content.