Two LoRAs, one clean layer-editing workflow: extract an object as a transparent layer, then remove it from the original image. 🤖 modelscope.ai/models/DiffSyn… 🤖 modelscope.ai/models/DiffSyn… ✂️ LayerExtract isolates a prompt-specified subject and exports it on a transparent background for independent editing. ✨ LayerRemove takes the source image and a reference layer, removes the corresponding object, and reconstructs the scene behind it. ⚡ Both are compact Qwen-Image-2.1 LoRAs that can be hot-swapped within the same DiffSynth-Studio pipeline. 🧩 Together they turn flattened images into editable assets for compositing, repositioning, replacement, and design workflows. 📜 LoRA weights: Apache 2.0. Qwen-Image-2.1 base-model terms also apply.

Jina-OCR-v1 parses entire pages into structured Markdown at 2.57 pages per second.🚀 🤖 modelscope.ai/models/jinaai/… 📄 modelscope.ai/papers/2609.03… 🏆 Scores 91.14 on OmniDocBench v1.6 and 83.4 on olmOCR-Bench, 7.4 points above DeepSeek-OCR on the latter. ⚡ Delivers the highest throughput among 14 evaluated systems at concurrency 32. 📑 Preserves text, formulas, tables, and reading order in a single pass. 🧠 The 3.4B MoE activates only 570M parameters per token, while FastMTP accelerates decoding without changing the autoregressive output. 📜 CC BY-NC 4.0. Commercial use requires permission.

NeoHorse-Jev-4B is now open—a compact model built to turn application states directly into structured decisions and probabilities. 🤖 modelscope.ai/collections/To… 🏆 Scores 77.70 across six text decision benchmark groups, ranking first among the four open-weight models with complete results in the comparison. ⚡ Prefill-only inference avoids autoregressive answer generation and supports three decision primitives: Choice for action selection, Noul for yes/no judgment, and Score for ordered ratings. 🎮 The same interface can power request routing, tool selection, workflow control, games, robot manipulation, and autonomous-driving simulations. 🖼️ Supports text inputs or a single image combined with text, returning normalized probabilities that applications can use directly. 🛠️ Deploy locally through vLLM, SGLang, Python, CLI, or HTTP. Apache 2.0.

Eight controls and inpainting in one checkpoint. Qwen-Image-2.1-Fun-Controlnet-Union adds precise structural guidance to Qwen-Image 2.1. 🎨 🤖 modelscope.ai/models/pai/Qwe… 🎛️ One control branch supports Canny, Depth, Grayscale, HED, Lineart, MLSD, Pose, and Scribble with no checkpoint switching. 🖌️ Control and inpainting share the same branch, allowing masked regions to follow both the prompt and a structural reference. 🧠 Sixteen injection points guide every second Transformer block while keeping the base model frozen. ⚡ CFG-distilled sampling runs at guidance scale 1.0, while prefix KV caching accelerates repeated denoising steps. 📜 Qwen Research License. Qwen-Image 2.1 base weights are required.

NVIDIA Nemotron 3 Diarization is now available on ModelScope—adding live speaker attribution to existing ASR workflows without replacing the transcription model. 🎙️ 🤖 modelscope.ai/models/nv-comm… 👥 Processes streaming audio and returns speaker labels and timestamps for up to eight speaker slots in a single conversation. ⚡ Its end-to-end streaming architecture avoids separately combining voice activity detection, speaker embeddings, clustering, and post-processing. 🧠 The 99.2M-parameter model uses a 31-layer Transformer encoder with RoPE and builds on NVIDIA’s Streaming Sortformer architecture. 🔌 Pair it with Nemotron ASR, Parakeet, Canary, Whisper, or another ASR system to create speaker-attributed transcripts. 🏢 Designed for meetings, contact centers, clinical conversations, live captioning, media analysis, and multi-party voice agents. 🖥️ Supports NVIDIA Ampere, Hopper, and Blackwell GPUs, with inference through NeMo Speech C++.

Digital PDFs or warped phone photos, TeleOCR parses them with one lightweight 1.2B vision-language model. 📜 Apache 2.0 License. 🤖 modelscope.ai/models/XingChe… 📃 modelscope.ai/papers/2608.12… 🏆 Scores 96.87 overall on OmniDocBench v1.6, the highest among the listed specialized VLMs, and ranks #1 in the ICDAR 2026 Sci-ImageMiner Challenge. 📷 Handles digital, photographed, curved, and degraded documents directly, without a separate dewarping model. 🧠 Combines geometry-aware synthesis, consensus-generated labels, image-based self-verification, and progressive training from vision-language alignment to reinforcement learning. ⚡ Supports structured parsing of text, tables, formulas, layouts, and reading order, with synchronous or asynchronous vLLM inference.

Xiaomi MiMo-V2.6 is now open—a native multimodal agent family built for large-scale reinforcement learning. 🚀📜 MIT License. 🤖 modelscope.cn/collections/Xi… 🏆 MiMo-V2.6-Pro scores 46 on the Artificial Analysis Intelligence Index and reaches 71.9 on DeepSWE v1.1, 89.9 on Terminal-Bench 2.1, and 82.0 on OSWorld-Verified. 🧠 The 1.02T MoE activates 42B parameters and supports text, images, video, audio, and a 1M-token context. ⚙️ One mixed RL run trains coding, general, visual, and cybersecurity agents together. Pro and Flash completed 30 steps each in under six days, producing around 750K trajectories. 🌐 The models support computer use, 3D creation, embodied control, coding, design, video, and music workflows.

Shanghai AI Lab and SJTU’s LUMIA Lab release NCP-ArchPreview, an open-weight 8.9B language modell. 📜 Apache 2.0. 🤖 modelscope.ai/collections/Sh… 📄 modelscope.ai/papers/2609.10… ⚡ Trained on 5.73T Dolma 3 tokens, it reaches OLMo-3-7B’s final Stage 1 loss with only 51.3% of the tokens, a 1.95× convergence gain. 🏆 Its Stage 1 macro-average rises from 46.59 to 49.04, with +5.99 on GSM8K and +4.28 on HumanEval. 🧠 NCP jointly predicts tokens and a concept sequence at one-quarter the length, then feeds those concepts back to guide generation. 🛠 Domain adaptation updates only the 17M-parameter concept module while keeping the token backbone frozen. 🚀 Concept-conditioned drafting improves mean accepted length by 4.17% with negligible overhead.

inclusionAI open-sources the Ming-Image-0.1-Design family, two complementary 6B models for visual-design workflows.📜 MIT License. 🤖 modelscope.ai/models/inclusi… 🤖 modelscope.ai/models/inclusi… 🏆 Design ranks #1 among open-weight models on the Artificial Analysis UI/UX Design leaderboard. Layer leads all 12 evaluated Crello settings and runs 4.3× faster than the evaluated 20B open-weight baseline. 🎨 Design generates complete UIs, dashboards, infographics, and posters at up to 2048×2048, with strong text rendering and native transparent RGBA output. 🧩 Layer decomposes flattened graphics into independently editable RGBA layers while preserving the original aspect ratio. 🧠 The pipeline combines multimodal prompt conditioning, a diffusion transformer, and a 4-channel VAE. Layer adds multi-frame generation for separate layer outputs.

Apsara 2026 is ON 🔥 Come find the ModelScope booth, snap a pic, and take home some merch: 🎒 Backpacks · Crossbody bags · Tote bags 🛋️ Neck pillows · ☕ Mugs · 💻 Laptop stands 🧸 Plush pendants · 🧢 Hats · 🧲 Fridge magnets 🎧 Earphone pouches · 📱 Phone chains ...and more. 📍 杭州国际博览中心二期 · 1F · 智能馆 · 1-5C · 魔搭社区 Hangzhou International Expo Center Phase II · 1F · Intelligence Engine · Booth 1-5C · ModelScope Come early before they're gone 👀

Qwen introduces RecreationBench, a benchmark for Hybrid Computer-Use Agents with 250 application-recreation tasks across Ubuntu, macOS, Windows, Android, and Web, spanning domains such as productivity, development, graphics, multimedia, and science. Unlike GUI-only or terminal-only benchmarks, agents must explore a running reference app, recreate it in code, and pass both programmatic tests and VLM-based visual evaluation. The playground is ready. Let’s build! 🚀Dataset: modelscope.ai/datasets/Qwen/…

NetEase-Youdao releases Confucius4-R2T2, a true-streaming ASR model for live captions, simultaneous translation, and voice agents. 🤖 modelscope.ai/models/netease… 🏆 R2T2 achieves SOTA latency and recognition quality among the evaluated open-source models, while remaining competitive with leading closed-source systems. ⚡ Configurable 80ms–2s chunks deliver 200–600ms average latency with near-offline recognition accuracy. 📝 Append-only decoding commits stable text without revising earlier words, avoiding transcript flicker and giving downstream agents reliable input. 🌍 Optimized for Chinese and English, with multilingual recognition, hotword prompts, and contextual prompts. 🧰 The GitHub repo includes inference code, a minimal example, and a vLLM backend for both offline and real-time streaming ASR. 📜 Code: Apache 2.0. Weights: NetEase Model Use License Agreement.