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.
LatestQwen · Alibaba3.8
Qwen · AlibabaVision-Language

Qwen3.8-27B Brings Vision to a Dense Model

Alibaba's latest Qwen release pairs image understanding with text under a permissive Apache-2.0 license.

Aug 5, 2026
Major releaseApache 2.0
Qwen3.8-27B

Alibaba's Qwen team has published Qwen3.8-27B, a 27-billion-parameter vision-language model that handles both conversational text and image understanding. It arrives under the permissive Apache-2.0 license, making it broadly usable in commercial and research settings.

Unlike the mixture-of-experts designs that have dominated recent large releases, Qwen3.8-27B is a dense model. That means all of its parameters are active on every forward pass — a simpler architecture that tends to be more predictable to serve and fine-tune, even if it trades away some of the efficiency gains MoE offers at larger scales.

What it offers

  • Dense 27B parameters, positioned in the mid-size tier between smaller on-device models and frontier-scale systems
  • Multimodal input: image-text-to-text alongside standard text chat
  • Apache-2.0 licensing, which permits redistribution and commercial use

The combination is notable because open, permissively licensed vision-language models in this size range remain relatively scarce. A 27B dense VLM is large enough to be capable at document, chart, and image reasoning, yet small enough to run on a single high-memory GPU with quantization.

As with prior Qwen drops, the weights are available directly on Hugging Face. Details on context length and benchmark performance were not specified in the release record, so teams evaluating it should confirm those specifics against the model card before committing to production use.

Sources

  • Qwen/Qwen3.8-27B

    Hugging Face

    Visit
OlderDeepgrove's Maple Preview bets on ternary-weight MoEDeepgrove · Reasoning · 4 months agoNewerMameLoshnLM Brings Yiddish Into the Open LLM EraOpenMOSS · Text / LLM · 4 months ago

Get the model

Hugging Face

Specs

Parameters27B
Context window262K tokens
Size55.6 GB
PrecisionBF16
ArchitectureQwen3_5ForConditionalGeneration
LicenseAPACHE-2.0
Downloads3.8M
Likes3.2K

Can you run it?

Runs on a 24 GB GPU at 4-bit.

  • BF16 (as published)

    80 GB GPU (A100 / H100) · 128 GB Mac Studio

    58.7 GB
  • 8-bit

    48 GB GPU (RTX 6000) or 2×24 GB · 64 GB Mac

    32.6 GB
  • 4-bit

    24 GB GPU (RTX 3090 / 4090) · 32 GB Mac

    19.8 GB
Your machine
GGUF builds
  • 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

Vision-LanguageText / LLM
10 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 Qwen · Alibaba

All Qwen · Alibaba releases →
Qwen · Alibaba/Speech → Text

Qwen-Audio-3.1-ASR Adds Speaker-Aware Transcription

Alibaba's new open audio-language model transcribes and diarizes speech across 30 languages under an Apache 2.0 license.

Oct 19, 2026
Qwen-Image-2.1-Turbo
Qwen · Alibaba/Text → ImageRuns on a laptop

Qwen-Image-2.1-Turbo targets faster generation

Alibaba's Qwen team ships a distilled, few-step variant of its image model built for speed.

Oct 9, 2026
Qwen-Image-2.1
Qwen · Alibaba/Text → ImageRuns on a laptop

Qwen-Image-2.1 Adds RGBA to Image Generation

Alibaba's Qwen team updates its open image model with transparency support and joint text-to-image and editing capabilities.

Sep 14, 2026

More in Vision-Language

All Vision-Language →
SenseNova-Loop-9B
SenseTime/Vision-LanguageRuns on a laptop

SenseTime Releases Looped 9B Multimodal Model

The company's new SenseNova-Loop-9B pairs vision-language capabilities with a recurrent looped-transformer design.

Nov 17, 2026
Hunyuan Hy4-flash
Tencent/Vision-LanguageMulti-GPU server

Tencent releases Hunyuan Hy4-flash multimodal MoE

The lightweight, mixture-of-experts model adds vision and text instruction-following to Tencent's open Hunyuan lineup.

Nov 17, 2026
Mirage
Cactus Compute/Vision-Language

Cactus Compute's Mirage brings vision tasks on-device

The new open vision-language model targets detection, segmentation, and visual Q&A without relying on the cloud.

Nov 5, 2026