ModelScope (@ModelScope2022) header
ModelScope (@ModelScope2022) profile picture

Driving innovations with open communities. 💬 Join our Discord: https://teapawt.net/t.co/4M9AHVh5qa

HangZhou, China
Joined April 2024
Photo by ModelScope (@ModelScope2022)Photo by ModelScope (@ModelScope2022)Photo by ModelScope (@ModelScope2022)Photo by ModelScope (@ModelScope2022)Photo by ModelScope (@ModelScope2022)Photo by ModelScope (@ModelScope2022)Photo by ModelScope (@ModelScope2022)Photo by ModelScope (@ModelScope2022)Photo by ModelScope (@ModelScope2022)Photo by ModelScope (@ModelScope2022)
LongCat-Video-Avatar 1.6 is out, pushing audio-driven avatar video past 10 minutes with more natural speech motion and stronger long-form consistency. 📜 MIT License. 🤖 modelscope.ai/models/meituan… 🏆 Ranks #2 in human preference: It places second among six evaluated systems in a 33,051-vote Arena, ahead of LongCat-Video-Avatar 1.5, HeyGen, OmniHuman-1.5, and Kling Avatar 2.0. 🗣️ More natural talking avatars: The update improves lip-audio sync, facial and body motion, and reduces exaggerated mouth artifacts during speech. 🎬 Built for long-form video: It reduces error accumulation during autoregressive continuation to preserve identity, visual quality, and temporal coherence beyond 10 minutes. 🧩 More than a single portrait workflow: Supports single- and multi-character audio-driven video generation, including audio-text-to-video, audio-image-to-video, and video continuation. ⚡ Optimized for serving: Includes end-to-end inference improvements, task-specific LoRAs, and an optional weight-only INT8 DiT checkpoint.
2
2
22
1,701
Try a live, real-time voice conversation with Haru-4B on ModelScope. Demo 👉 modelscope.ai/studios/AI-Mod… Upload or record a 3-second to 3-minute voice reference, set a character and conversation language, then start talking. Hold the live mic button to speak and follow the exchange with real-time captions. Haru is designed to keep the conversation flowing with natural backchannels and interruptions, rather than making you wait for each turn to end. Headphones recommended. Each session supports up to five minutes of conversation.
3
37
2,344
What if your research agent didn’t stop at the first answer? Built by @SolenopsisCampo , Venastine Research Harness is now available to try on ModelScope. Its ten-pass Research mode drafts an answer, checks claims against sources, challenges assumptions, and revises only what gets flagged. Use Chat for everyday tool-assisted questions, or Research for work that needs a more auditable trail—all in a terminal-style workspace in your browser. 🤖 modelscope.ai/studios/Venast…
1
5
978
Edge0-145B brings 145B-scale agent capabilities to a single consumer GPU, with 10.12B active parameters per token and 256K context. 🤖 modelscope.ai/collections/Ed… 🏆Strong agent results: In Edge0’s reported comparison, the 145B model tops the listed models on Terminal-Bench 2.1 with 86.5, while reaching 64.6 on DeepSWE 1.1. ⚡ 145B, running locally: The Q3_K build reports 121.2 tok/s for agent decoding and 3,554.9 tok/s for 16K-token prefill on an RTX 5090 D 24GB. 🧠 Sparse by design: Its latent MoE architecture activates just 10.12B parameters per token across 512 experts, making a much larger model practical for local deployment. 🛠️ Pick the deployment fit: Edge0-145B prioritizes quality, while Edge0-120B offers a faster 8.64B-active alternative. Both are available in Q3_K and NVFP4 builds through Edge0’s dedicated local runtime. 📜 Edge0 Community Model License 1.0 applies to Edge0 contributions; upstream Qwen materials remain under the Qwen Community License 1.0.
4
4
78
3,921
RoboValue-Dataset is now on ModelScope: fine-grained training and evaluation data for robotic value models. 🤖 modelscope.ai/datasets/RoboV… 📦 Sim and real manipulation: 3,535 expert episodes across 15 simulation and 20 real-world tasks. 🧠 Evaluate execution, not just outcomes: Covers task state, progress, failure recovery, and value consistency. 🔍 Stress-test value judgments: Counterfactual instructions, repeated states, recovery trajectories, and alternative task orders expose unreliable feedback. 🔄 Test generalization: Includes cross-embodiment and cross-environment settings, with Full-Shot and One-Shot subsets in LeRobot v3.0 format. 📜 Code, scripts, and documentation are released under the MIT License.
3
28
2,210
Voder is now open source. Built by the VeriLoop team and supported by the Intelligent Robotics Laboratory at Tsinghua SIGS, it’s a coding agent that verifies proposed changes before writing them to the workspace. 🛡️ Independent checks before write-back. With VeriLoop-Governed Recurrence (VGR), the agent proposes and revises changes. A Rust governance layer authorizes the write-back only after the exact proposed change passes independent checks. The team is exploring how this approach could make recursive self-improvement safer. 📊 Locally audited benchmarks with DeepSeek V4.1 Flash 1017 • DeepSWE v1.1 · 78.8% (89/113) • Terminal-Bench 3.0 · 41.9% (31/74) • Terminal-Bench 4.0 · 45.5% (30/66) • SWE-Marathon · 45.0% (9/20), with reference assistance These are the team’s local audit results. Voder is available on GitHub and the VS Code Marketplace. 📄 modelscope.ai/posts/434415 💻 github.com/brucewang12345678…
4
2
14
2,542
🚀 MiniMax-H3-Image is now live on ModelScope Civision! ✍ Turn words into visuals 🖼 Mix & edit with up to 8 reference images ✨ Bring every detail to life in native 4K One idea. Endless possibilities. 🔗 Create something amazing: modelscope.ai/civision/image…
1
12
2,063
Unsloth reports 5-10% gains across most tested benchmarks with Dynamic V3.1 (preview), its latest quantization update for the open-weight Qwen3.8-27B GGUFs. 📜 Apache-2.0 License. 🤖 modelscope.ai/models/unsloth… 🧪 Evaluated on practical workloads: coding, agents and chat, with benchmarks including Terminal Bench and DeepSWE. 🎛️ Pick your memory budget: most quantizations from 2-bit to 6-bit are available. The 3-bit files retain UD3 because V3.1 did not deliver enough improvement to justify replacing them. 💻 Run locally: the 4-bit version targets roughly 16-19GB VRAM, or a Mac with 24GB memory. Actual requirements depend on context length and offloading. 🛠️ Choose your stack: run through Unsloth Desktop, serve via unsloth run, or use llama.cpp. 📊 Benchmark details matter: Unsloth's analysis examines embedding-size accounting, context budgets and repeated runs, rather than relying on a single score.
2
25
2,965
The Sophon-Realtime demo is now available on ModelScope Studio. Try the latest version live! modelscope.ai/studios/Advent…
Sophon-Realtime v1.1 from Adventists AI @BKJinAI adds zero-shot voice cloning to an open-weights, full-duplex voice model. Drop a few seconds of someone's voice into model/voices/, set "voice" to its name, and talk to it in that voice in real time. No fine-tuning, no restart, one GPU. Barge-in and laughs still work, and a voice cloned from a clip in one language can speak another. Available in the Tag and Flash builds on ModelScope. Live mode is a drop-in replacement for the OpenAI GPT-Live API, and an E2E build with 0.21 s to first audio is coming soon. Please only clone voices you have permission to use. 🤖modelscope.ai/models/Adventi…
1
16
2,041
Deploy MiniMax-H3-Image on ANY hardware with ultra-low cost! 🤖 modelscope.ai/models/DiffSyn… Fast, low-cost inference is now possible with EntroPack. Choose your checkpoint: 🔹 80GB VRAM: bf16_table_pruned.safetensors (Fully lossless pruned model) 🔹 24GB VRAM: dit_8bit_int8.safetensors (Near-lossless INT8 + ConvRot quantization) 🔹 8GB VRAM: dit_extreme_3bpp.safetensors (Extreme 3-bit quantized model) ComfyUI workflows also available in the model repository. Try it now!
1
2
32
3,364
Switch from Claude Code to Codex, Cursor or your own harness, and Nex-N2.6 still calls only the tools that are actually there. The open-weight 321B multimodal agentic model from Nex-AGI is built on GLM-5.3-Flash for computer use, coding and long-horizon tasks. 📜 Apache 2.0 🤖 modelscope.ai/models/nex-agi… 🛠️ Harness-agnostic tool use: 100% valid tool selection across 956 real-world contexts, matching GPT-6-Sol and up from 80.56% for Nex-N2.5-Pro. 🔁 Far less looping: reasoning repetition drops 90%, from 16.0% to 1.6%, on 698 real failure cases replayed from the exact point where the previous model stalled. Fewer wasted tokens on long runs. 🖥️ Stronger computer use: 83.0 on OSWorld, 5.0 points above its GLM-5.3-Flash base, while holding 84.1 on Terminal-Bench 2.1. ⚡ Ready to run: 256K context via API, or self-host the FP8 weights on SGLang behind an OpenAI-compatible endpoint with low, high or max reasoning effort.
9
2
2
66
5,205
Sophon-Realtime v1.1 from Adventists AI @BKJinAI adds zero-shot voice cloning to an open-weights, full-duplex voice model. Drop a few seconds of someone's voice into model/voices/, set "voice" to its name, and talk to it in that voice in real time. No fine-tuning, no restart, one GPU. Barge-in and laughs still work, and a voice cloned from a clip in one language can speak another. Available in the Tag and Flash builds on ModelScope. Live mode is a drop-in replacement for the OpenAI GPT-Live API, and an E2E build with 0.21 s to first audio is coming soon. Please only clone voices you have permission to use. 🤖modelscope.ai/models/Adventi…
1
2
4
53
5,995
4K x 4K in 10-20 seconds. MiniMax releases H3-Image, an open-weight model for high-resolution image generation and editing. • Precise control: open-model SOTA per MiniMax in instruction following, with edits and compositions from up to 16 reference images. • Photoreal by default: rich detail and natural texture, built to avoid the glossy "AI look." • Flexible canvases at full 16MP: 4096x4096 up to 6272x2688 (21:9), generated directly with no upscaler. • Text that reasons: renders logic puzzles in Chinese and English and solves them correctly on the board. • Built to serve: H3's single-stream DiT minus the audio branch, distilled to 6 steps with no CFG. Runs on SGLang, vLLM, diffusers and ComfyUI. 🤖 modelscope.ai/models/MiniMax… 🤖 Prompt expander: modelscope.ai/models/MiniMax…
2
10
1
77
4,935
LingBot-Touch has landed: an open-source tactile foundation model for transferring touch-based skills across robot hands, sensors, and manipulation tasks. 📜 Apache-2.0 License. 🤖 modelscope.ai/collections/Ro… 🖐️ Pretrained on touch at scale: Learns from 4,983 hours of tactile recordings, 1.33B frames, and 464K interaction episodes across piezoresistive, magnetic, and visuotactile sensors. 📊 Built for dexterous manipulation: Evaluated across 26 real-world tasks and 12 simulated RL tasks, including fragile-object handling, deformable objects, and bimanual coordination. 🔄 Transfer across sensors and hands: A shared Haptic Transformer maps heterogeneous tactile signals into one reusable representation, enabling transfer to unseen tactile sensors and real-world dexterous hands. ⚡ High-frequency and lightweight: The HaT-Tiny encoder uses just 5.46M parameters while supporting responsive downstream tactile inference. 🛠️ Ready to build with: Tiny and Small checkpoints, inference code, sensor interfaces, and sample data are released for three tactile modalities.
1
26
1,600
Meituan has released the weights for LongCat-2.5 under the MIT License. The native multimodal MoE is built for long-horizon coding and agent tasks, with 1.6T total parameters and about 48B activated per token. 🤖 modelscope.ai/models/meituan… • Stronger coding agents. LongCat-2.5 improves substantially over LongCat-2.0 on DeepSWE and Terminal-Bench v4.0 in Meituan’s tests. • Visual coding. The model processes image and text tokens together, so it can use a screenshot as a reference when writing webpage code. • 3D modeling through code. The model can turn a text description into a Blender script and leads the models in Meituan’s 3DCodeBench Text-to-3D comparison. • Training for longer tasks. Agentic RL uses real engineering tasks and reproducible agent failures. LongCat-2.5 also posts the highest BrowseComp score in Meituan’s reported comparison. • INT4 inference on NPUs. With quantized KV cache, W4A4 nearly doubles decode throughput over W8A8KV16 in Meituan’s stress test.
1
11
1,437
Adventists AI just open-sourced Sophon-Realtime v1 on ModelScope: a full-duplex voice model that hears how you speak, and answers with tone of its own. • It keeps listening while you talk, answers ~0.4 s after you stop, and stops the moment you cut in • It laughs, sighs and clears its throat when the conversation calls for it • #1 among open full-duplex models on VoiceBench (82.8) and Big Bench Audio (81.8%) • 48% Pass@1 on Full-Duplex-Bench v3 tool calling, ahead of Grok (43%) Apache-2.0 weights. Built by @BKJinAI and team. modelscope.ai/collections/Ad… modelscope.ai/studios/Advent…
19
14
2
133
7,467
🌏 Bringing Open-Source AI to the Geoscience Community at IEEE AP-GARSS 2026! Together with the @IEEE_GRSS Beijing Student Chapter Branch, we organized the Open Source GeoAI Practice Challenge in Qingdao, bringing researchers and developers together to explore open-source AI for geoscience. 🎉 Congratulations to all eight finalist teams! From satellite image generation and AI agents to typhoon monitoring and wildlife conservation, every team turned innovative ideas into interactive ModelScope Studios you can try directly in your browser! 🚀 Explore the projects and prizes: modelscope.ai/events/119 Special thanks to Prof. Saibun Tjuatja, IEEE GRSS President, Prof. Xiaofeng Li, AP-GARSS 2026 General Chair, and all our judges, organizers, and volunteers for their support! Together, we're making GeoAI research more open, accessible, and practical. ❤️ #APGARSS2026 #IEEEGRSS #ModelScope #GeoAI #OpenSource #RemoteSensing
2
9
1,559
• 1st Prize: GenEarth: Open Generative Foundation Models for Global-Scale, Text-Driven Earth Observation modelscope.ai/studios/KeyanC… • 2nd Prize: RSMeM: Knowledge-Enhanced Memory Evolution for Remote Sensing Agents with Systematic Evaluation modelscope.ai/studios/benson… • 2nd Prize: PhysEarth-Agent: Agentic Physics-Based Modeling for Earth Science modelscope.ai/studios/yiwenf…
1
3
734
THX-02 from @halx_ai is open source: a 4.5B multimodal decision model trained on requests up to 32K tokens, with a 60M adapter for selective thinking. It returns probabilities over your supplied options. 🤖 modelscope.ai/models/HAL-X/T… 🧠 Reason selectively: Auto mode generates a short reasoning note when the direct answer is uncertain and an internal gate identifies a need for reasoning, then combines the direct and thought-assisted predictions. ⚡ Fast direct decisions: Median latency is 50 ms directly and 79 ms in auto mode on one RTX PRO 6000. Requests that trigger thinking typically take 0.3–1 second. 📊 Measured gains from thinking: Auto mode raises Decision Index 0.3 from 63.83 to 65.06 on a 6,018-request public-suite sample. Full-suite and private leaderboard evaluations remain pending. 🎯 Define your own decisions: Ask up to 64 questions per request using multiple-choice, yes/no, or ordered ratings. New labels need no retraining. 🖼️ Text, images, and charts: Accepts up to four images per request, with post-training and evaluation across more than 20 languages. 📜 Apache 2.0.
5
2
20
1,690
Baidu is opening ERNIE-5.5-Flash-Base straight out of pre-training: an open-weight foundation for domain adaptation, instruction tuning, and RL research. 🤖 modelscope.ai/models/baidu/E… • 280B total parameters; 14B activated per token • Pre-trained on 33.5T tokens, with training context up to 128K • Up to 1M-token context for fine-tuning and inference • FP8 and BF16 weights 🛠️ A starting point for your own training: The checkpoint precedes all mid- and post-training, leaving developers to choose their own adaptation methods and objectives. 📊 Strong base-model results: Scores 78.4 on MMLU-Pro, 82.3 on HumanEval+, and 40.1 on LiveCodeBench, ahead of the three compared base models on these benchmarks under a unified evaluation pipeline. 🧠 Compressed attention for long inputs: Combines 37 HCA layers, each caching one entry per 128 tokens, with six MLA-DSA layers for full-context access through sparse attention. ⚙️ A built-in speculative draft: ECHO reuses the backbone's final layer to predict future tokens, replacing a separate MTP module. 📦 Deployment: Transformers, SGLang, and vLLM-ERNIE paths are available. The Base model uses the completions endpoint.
2
1
30
2,076