The Open Weights
LatestModelsLeaderboardsCompanies
Subscribe
The Open Weights

The daily record of open-source AI. New model releases, leaderboards, and what's coming next — written for people who ship.

Refreshed every 12 hours

Discover

  • Latest releases
  • New today
  • Trending models

Browse

  • All models
  • Companies
  • Categories
  • Leaderboards

About

  • About
  • Editorial policy
  • Hardware estimates
  • RSS feed
  • llms.txt
  • Newsletter

© 2026 The Open Weights. An independent publication.

PrivacyTermsSMSAggregated by Claude · curated by humans.
LatestDeepSeekV3.2
DeepSeekText / LLM

DeepSeek-V3.2 Arrives With FP8 Weights, MIT License

The new Mixture-of-Experts model from DeepSeek AI combines an efficient FP8 architecture with a fully permissive license for commercial use.

Dec 1, 2025
NotableMIT
DeepSeek-V3.2

DeepSeek AI has released DeepSeek-V3.2, a new open-source language model that stands out for its technical efficiency and permissive licensing. The model uses a Mixture-of-Experts (MoE) architecture, placing it among a class of powerful yet computationally lean models designed for scalable inference.

The most notable technical feature is its use of FP8 (8-bit floating point) weights. This precision format significantly reduces the model's memory footprint compared to traditional 16-bit models. For developers, this translates to faster inference and the ability to run the model on more accessible, consumer-grade hardware.

The MoE design complements this focus on efficiency. By routing input tokens to specialized "expert" sub-networks, the model can achieve the performance of a much larger dense model while only activating a fraction of its parameters for any given task, further reducing computational costs.

Perhaps most importantly for the open-source community, DeepSeek-V3.2 is available under the MIT License. This choice removes many of the commercial use restrictions seen with other high-performance models, offering developers and businesses a truly open foundation for building new applications.

Sources

  • deepseek-ai/DeepSeek-V3.2

    Hugging Face

    Visit
OlderFlashLabs Releases Chroma-4B, an Any-to-Any ModelFlashLabs · Any-to-Any · 12 months agoNewerResemble AI Releases Chatterbox Turbo for Open TTSResemble AI · Text → Speech · 11 months ago

Get the model

Hugging Face

Specs

Context window164K tokens
Size689.3 GB
PrecisionFP8
ArchitectureDeepseekV32ForCausalLM
LicenseMIT
Downloads8.9M
Likes1.6K

Can you run it?

Datacenter only — about 412.6 GB even at 4-bit.

  • FP8 (as published)

    Multi-node cluster

    690.9 GB
  • 4-bit

    8×H100 80 GB node

    412.6 GB
Your machine
GGUF builds
  • Mixture-of-experts: every expert must be loaded, so memory follows total parameters, not the active slice.
  • KV cache sized for 8,192 tokens of context; longer prompts need proportionally more.

Based on Hugging Face weights metadata. Assumes an 8K context and ~1 GB runtime overhead; actual needs vary. How we estimate


Modalities

Text / LLM
2 versions — view changelog

The Weekly Weights

Every open release that mattered, one email a week.

0 comments

No comments yet. Be the first to weigh in.

More from DeepSeek

All DeepSeek releases →
DeepSeek-V4.1-Flash
DeepSeek/Text / LLMDatacenter

DeepSeek Releases V4.1-Flash, an Open Multimodal MoE

The new mixture-of-experts model handles text, vision, and reasoning, and ships under a permissive MIT license.

Oct 17, 2026
DeepSeek-V4.1-Flash (1017)
DeepSeek/Vision-LanguageDatacenter

DeepSeek V4.1-Flash Adds Vision and Reasoning

The new MIT-licensed checkpoint folds image understanding and reasoning into DeepSeek's fast, mixture-of-experts Flash line.

Oct 17, 2026
DeepSeek-V4-Flash-Vision-Exp
DeepSeek/Vision-LanguageMulti-GPU server

DeepSeek adds vision to its V4 Flash line

An experimental, MIT-licensed vision-language model brings image understanding to DeepSeek's fast V4 Flash architecture.

Aug 31, 2026

More in Text / LLM

All Text / LLM →
Meta AI/Text / LLM

Meta AI debuts Muse Ember, a compact open model

The company's smallest, most efficient open-weight language model targets lean, on-device text workloads.

Nov 10, 2026
Inkling-2-Small
Thinkingmachines/Vision-LanguageMulti-GPU server

Thinking Machines ships Inkling-2-Small, an open MoE VLM

The Apache-2.0 vision-language model uses a mixture-of-experts design and lands openly on Hugging Face.

Nov 4, 2026
LongCat 2.5
Meituan/Text / LLMDatacenter

Meituan Releases LongCat 2.5, an MIT-Licensed MoE

The Chinese delivery giant expands its open-model push with a multimodal, agentic mixture-of-experts system.

Nov 4, 2026