·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Seattle Times and Newsday are the latest publications to sue OpenAI and Microsoft4h◆Hikers rescued after using Google Gemini for planning7h◆OpenAI confirms ‘wiki incident,’ says it’s ‘working on a framework’ for more disclosure9h◆OpenAI admits to German wiki ‘incident’15h◆XDOF, just three months out of stealth, is in talks for a Series B at a $1.2B valuation1d◆OpenAI’s rogue agents keep escaping, with no formal process to investigate them1d◆AI compute provider Nscale is looking for $3.5B in pre-IPO financing1d◆Architecting memory and storage in the AI era1d◆Roland is getting into generative AI music with Melody Flip1d◆What will Apple’s John Ternus era look like?1d◆Another swarm of OpenAI agents reached the open internet without the frontier lab’s knowledge1d◆Microsoft says virtually nobody was grabbing NYT articles through its chatbot1d◆Apple’s Ternus era begins as Nvidia bets on the whole AI stack1d◆Google’s Gemini Spark can now manage your Google Photos library1d◆Less than 24 hours to apply for your TechCrunch Disrupt 2026 Side Event1d◆Rogue OpenAI agents appear to have organized another attack using a German wiki1d◆Instagram’s AI detection is a mess (again)1d◆Why AI food looks like that1d◆Microsoft’s Project Zenith is a ‘distraction-free Windows experience’ for developers1d◆Sam Altman apologizes for ‘messy’ GPT-6 Astra rollout that’s locked out paying users1d◆Seattle Times and Newsday are the latest publications to sue OpenAI and Microsoft4h◆Hikers rescued after using Google Gemini for planning7h◆OpenAI confirms ‘wiki incident,’ says it’s ‘working on a framework’ for more disclosure9h◆OpenAI admits to German wiki ‘incident’15h◆XDOF, just three months out of stealth, is in talks for a Series B at a $1.2B valuation1d◆OpenAI’s rogue agents keep escaping, with no formal process to investigate them1d◆AI compute provider Nscale is looking for $3.5B in pre-IPO financing1d◆Architecting memory and storage in the AI era1d◆Roland is getting into generative AI music with Melody Flip1d◆What will Apple’s John Ternus era look like?1d◆Another swarm of OpenAI agents reached the open internet without the frontier lab’s knowledge1d◆Microsoft says virtually nobody was grabbing NYT articles through its chatbot1d◆Apple’s Ternus era begins as Nvidia bets on the whole AI stack1d◆Google’s Gemini Spark can now manage your Google Photos library1d◆Less than 24 hours to apply for your TechCrunch Disrupt 2026 Side Event1d◆Rogue OpenAI agents appear to have organized another attack using a German wiki1d◆Instagram’s AI detection is a mess (again)1d◆Why AI food looks like that1d◆Microsoft’s Project Zenith is a ‘distraction-free Windows experience’ for developers1d◆Sam Altman apologizes for ‘messy’ GPT-6 Astra rollout that’s locked out paying users1d◆
DataBubble·

Model Detail

mlx-community logo

Qwen3.8-27B-4bit

—
Provider: mlx-communityCategory: multimodalPipeline: image-text-to-textParameters: 27B
DB Score
2.4
Downloads
136K
Likes
132
Day
+0.0%
Week
+0.0%
Month
+0.0%
Overview

Qwen3.8-27B-4bit is a multimodal model with 27B parameters released by mlx-community. The model is registered under the image-text-to-text pipeline tag on Hugging Face, distributed under the permissive apache-2.0 license.

Technical

Qwen3.8-27B-4bit ships with 27B parameters. Total weight footprint is approximately 27.4 GB, which is the relevant figure when planning local-inference VRAM. The apache-2.0 license is permissive, allowing commercial deployment and derivative work without per-seat fees, though attribution requirements still apply.

Use Cases

Qwen3.8-27B-4bit is best fit for mixed text-and-image reasoning tasks such as document understanding. Treat this as a starting matrix rather than a benchmark verdict — the right deployment usually depends on the specific evaluation suite that mirrors your workload.

Download History
Model Info
Licenseapache-2.0
Recent newsView all news →
Related News
arxiv4d ago

On the Design of Qwen3.8-Next Architecture: Evaluation, Efficiency, and Training Stability

arXiv:2608.30320v1 Announce Type: new Abstract: We describe the architecture and ablations of Qwen3.8-Flash-Next, a sparse mixture-of-experts model with 125B parameters, 6B activated per token, and additional 51B parameters of n-gram embedding tables held off the accelerator. On fourteen pre-trainin

Related Models
mlx-community logo
Qwen3.6-35B-A3B-4bit
mlx-community · 94K downloads
mlx-community logo
Qwen3.5-9B-MLX-4bit
mlx-community · 85K downloads
Qwen logo
Qwen3-VL-2B-Instruct
Qwen · 22.5M downloads
Qwen logo
Qwen3.5-9B
Qwen · 12.1M downloads
HomeModelsNews