Alibaba just introduced Prompt Relay, a training-free technique that lets video models jump between radically different scenes without messy visual bleeding. By routing time-stamped prompts directly inside cross-attention layers, it keeps each action strictly in its designated slot while smoothing the handoff. The implementation details are up, with full code rolling out shortly.
Alibaba just launched Wan2.7-Image, a unified model that combines image generation, editing, and understanding. It renders print-quality text across 12 languages, offers precise facial and HEX color controls, and outputs up to 12 consistent images per prompt, solving a major headache for multi-panel creators.
Alibaba just rolled out Wan 2.7 Image, a unified model that handles both generation and precision editing. It focuses on lifelike human faces with natural variation, lets creators lock brand palettes using up to eight hex codes, and renders multi-language text up to 3,000 tokens. Why it matters: Pumping out up to 12 consistent shots in a single prompt makes storyboards and marketing assets far easier to build.
Alibaba just open-sourced PrismAudio, a lightweight 518M-parameter model that generates synchronized sound effects for silent videos. By combining reinforcement learning with chain-of-thought reasoning, it generates 9 seconds of audio in just 0.63 seconds, outperforming multi-billion-parameter rivals in both speed and benchmarks.
Alibaba just open-sourced Fun-CineForge, a multimodal dubbing model that syncs speech directly to characters' lip movements and timestamps. By treating time as an explicit modality, it handles fast speaker switches and facial occlusions cleanly, turning automated film dubbing into a realistic studio workflow.
Alibaba just open-sourced CoPaw, a personal AI assistant featuring a web UI and multi-channel messaging support. It connects directly with DingTalk, Lark, Discord, and iMessage, running locally via Ollama or MLX with MCP modular tools. Basically, you get automated cross-channel workflows and daily summaries without writing a single line of code.
Alibaba just dropped Mobile-Agent-v3.5 alongside the GUI-Owl-1.5 model family, spanning from 2B to 235B parameters. Built on Qwen3-VL, it natively unifies vision, reasoning, and tool execution across desktop, mobile, and browser environments, topping 20+ benchmarks. Model weights are now open on Hugging Face with live demos available.
Alibaba just open-sourced DeepGen 1.0, a 5B unified model that handles both text-to-image generation and complex visual editing. Beyond modifying in-image text and shifting perspectives, it can even solve mazes, outperforming much larger models on editing benchmarks. The catch is its 72GB footprint on Hugging Face, putting consumer-grade hardware to the test.
Alibaba's Tongyi Lab just open-sourced Z-Image, the undistilled foundation model behind its fast Turbo edition. Built specifically for customization, it fixes the Turbo version's repetitive faces and compositions while adding negative prompt support. It needs 50 steps instead of 28, and ComfyUI already supports it out of the box.
Alibaba just released Qwen3-TTS, an open-source speech model that clones any voice from just seconds of audio and lets you design new vocal styles using text prompts. Packing a dual-track architecture with 97ms end-to-end latency, it can fluidly generate cross-lingual speech and regional dialects for live streaming and AI agents.
Researchers just open-sourced CoDance, a new animation model that can animate multiple subjects from a single, potentially misaligned pose sequence. It breaks rigid spatial constraints and accurately rebinds motion to target characters using text prompts and masks. All code and model weights are now publicly available.
Alibaba researchers just introduced Thinking with Map, an agentic framework that equips vision-language models with map-browsing capabilities for image geolocalization. By combining reinforcement learning with parallel test-time scaling, the model achieved a 22.1% accuracy within 500 meters—outperforming Gemini-3-Pro's map-grounded mode by 2.8x.
Alibaba's open-source Qwen Code CLI just rolled out updates spanning v0.2.2 to v0.3.0, introducing Stream JSON input and output. That makes hooking the coding agent into CI/CD pipelines, SDKs, and automated workflows far cleaner, alongside full UI localization and a 20MB buffer guardrail against memory crashes.
Qwen just introduced Qwen-Image-Edit, a 20B-parameter foundation model built specifically for advanced image and text editing. It can directly modify Chinese and English text inside images while preserving fonts, alongside pulling off full 180-degree novel view synthesis and localized element tweaks. You can already try it inside Qwen Chat.
Alibaba just released Qwen-Image, an open-source 20B MMDiT foundation model built to tackle complex text rendering in both English and Chinese. It handles multi-line typography, paragraph layouts, and precision editing without garbling letters. It delivers native bilingual graphic design directly to the open-source image ecosystem.
Qwen just launched QwQ, a mid-sized reasoning model that matches heavyweight frontier systems like DeepSeek-R1. Alibaba found that scaling reinforcement learning yields massive leaps in math and coding, proving mid-sized models can punch way above their weight against massive MoE giants.
Researchers just introduced Animate Anyone 2, an upgraded image animation framework that finally solves character-environment interactions. Instead of producing floating figures pasted onto backgrounds, it extracts environmental cues and uses a dedicated object guider so animated characters can naturally touch, sit on, and interact with surrounding props. It is a huge leap toward realistic video generation with consistent physical affordance.
AI researcher Steven Hoi just joined Alibaba to lead its AI-to-Consumer business group. He plans to focus on building foundation models and AI agent applications for consumer markets over the coming decade. It shows Alibaba is getting serious about turning frontier model tech into everyday consumer software.
Alibaba just released Qwen2-VL, open-sourcing the 2B and 7B models under Apache 2.0 while offering its flagship 72B version via API. The model handles dynamic resolutions, parses 20-minute videos, and powers mobile device agents. Why it matters: The 72B flagship rivals GPT-4o on visual benchmarks, bringing top-tier multimodal power to open weights and developers.
Alibaba researchers just unveiled EMO, an expressive audio-driven portrait-video generation framework. Give it a single still image and an audio clip—talking or singing—and it animates the portrait with realistic facial expressions and head poses for any duration. It even keeps up with fast-paced rap and works across multiple languages, paintings, and 3D models.
