ModelScope

@ModelScope2022

Driving innovations with open communities. 💬 Join our Discord: discord.gg/hgKcfgXHAQ

HangZhou, China

modelscope.ai

Joined April 2024

Replies Articles Media

Pinned
ModelScope
@ModelScope2022
Aug 14
🚀Qwen3.8-27B is officially open-sourced! Apache 2.0 licensed: free to download, deploy, and use commercially. 🤖modelscope.ai/models/Qwen/Qw… Your new go-to model for local deployment is here: 🧠 Native multimodal dense architecture. Understands images, text, and video. 🎛️ Flexible
ModelScope
@ModelScope2022
4h
Better prompts, better edits. ✍️ Wuli-Art introduces Gemma-4-for-Qwen-Image-Edit-2511-Prompt-Extend, a prompt optimizer for Qwen-Image-Edit-2511. 🤖 modelscope.ai/models/Wuli-Ar… ✨Turns an image and a simple request into a clearer, more precise editing prompt. 🧠 Fine-tuned from
ModelScope
@ModelScope2022
Aug 14
The first open-weight dots3 model is here, with BF16 and deployment-friendly FP8 checkpoints. 🔴 dots studio releases dots3-note-prev for multimodal reasoning and agents. 🤖 modelscope.ai/collections/do… 🏆 Scores 81.4 on ARC-AGI-2 in the team’s evaluation. ⚡ 280B MoE with only 16B
ModelScope
@ModelScope2022
Aug 14
✨ DiffSynth-Studio turns MiniMax-H3 capabilities into reusable video effect templates. 🤖 modelscope.ai/models/DiffSyn… 🎬 Apply effects like Bullet Time, Dark Magic, Fire Breath, and Four Seasons, then mix them or build your own from text, images, videos, and training data. ⚡ The
00:00
ModelScope
@ModelScope2022
Aug 14
DeepSeek-V4-Pro is officially here. 🚀 Built for complex agentic and production workloads. 🤖 modelscope.ai/models/deepsee… 💻 Major agent upgrades, with strong gains in coding, tool use, and full-stack development. 🧠 Flexible reasoning effort: low for simple tasks, high for daily