# DeepSeek Refreshes V4-Flash With New 0731 Checkpoint

> The MIT-licensed mixture-of-experts model returns in an updated build shipping with FP8 weights for cheaper inference.

Published by The Open Weights on Jul 31, 2026. Canonical: https://theopenweights.com/news/deepseek-v4-flash-0731-s1rd

## Key facts

- Company: DeepSeek
- Model: DeepSeek-V4-Flash-0731
- Version: V4-Flash-0731
- Category: Text / LLM
- Modalities: Text / LLM, Reasoning
- License: MIT (Open weights, commercial use allowed)
- Context window: 1.0M tokens
- Architecture: DeepseekV4ForCausalLM
- Precision: INT8
- Weights size: 305.8 GB
- Significance: major
- Published: 2026-07-31
- Last verified: 2026-08-25
- Hugging Face: https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731
- Canonical URL: https://theopenweights.com/news/deepseek-v4-flash-0731-s1rd

DeepSeek has published a new checkpoint of its V4-Flash model, tagged 0731, on [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731). The release continues the company's practice of shipping open weights under a permissive MIT license, which allows commercial use and modification with minimal restrictions.

V4-Flash is a mixture-of-experts (MoE) model, an architecture that activates only a fraction of its total parameters for any given token. That design lets a model carry a large knowledge capacity while keeping the compute cost of each inference step comparatively low — a fitting approach for a checkpoint branded "Flash," a label the industry typically reserves for latency- and cost-sensitive workloads.

The checkpoint ships with FP8/8-bit weights, a quantized format that shrinks the memory footprint and can speed up serving on modern accelerators without the overhead of full-precision inference. For teams running their own infrastructure, that combination of open weights and reduced precision lowers the barrier to self-hosting.

## Why it matters

DeepSeek has become one of the more closely watched sources of openly licensed frontier-style models, and each new drop is scrutinized by developers weighing alternatives to closed APIs. A few practical notes on this release:

- It carries an **MIT license**, among the most permissive available.
- It uses a **mixture-of-experts** design aimed at efficiency.
- Weights are distributed in an **FP8/8-bit** format for lighter deployment.

As with any fresh checkpoint, the real test will come from independent evaluation. Full documentation and files are available on the model's [Hugging Face repository](https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731).

## Get the model

- [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731)

## Sources

- [deepseek-ai/DeepSeek-V4-Flash-0731](https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731) — Hugging Face, Jul 31, 2026

---
Source: The Open Weights (https://theopenweights.com/). Aggregated and written by Claude, curated by humans. Cite as: "DeepSeek Refreshes V4-Flash With New 0731 Checkpoint", The Open Weights, Jul 31, 2026, https://theopenweights.com/news/deepseek-v4-flash-0731-s1rd