Stitch might be the most underrated Gemini 2.5 use case so far. This isn’t a toy it’s a productivity multiplier.
MULTIMODAL AI
-
AI Week: Apple’s Gap, Music Startups, Emotional Intelligence
By
–
Top stories in AI today: – AI letdown expected at Apple’s WWDC
– Record giants, music AI startups eye deals
– Transform deep research into visual infographics
– AI beats humans in emotional intelligence
– 4 new AI tools & 4 job opportunities Read more: https://
therundown.ai/p/apples-ai-ga
p-year
… -
Google Launches Med-Gemma Open Source Medical AI Model
By
–
Google vient de lancer Med-Gemma : une IA médicale open source, capable d’analyser des radios, scanner des tumeurs et expliquer ses diagnostics en langage clair.
— VISION IA (@vision_ia) 2 juin 2025
Modèle léger. Téléchargeable. Multimodal.
Un médecin dans votre poche.
Extrait de ma vidéo sur les outils IA : pic.twitter.com/jre5RWPMmaGoogle vient de lancer Med-Gemma : une IA médicale open source, capable d’analyser des radios, scanner des tumeurs et expliquer ses diagnostics en langage clair. Modèle léger. Téléchargeable. Multimodal. Un médecin dans votre poche. Extrait de ma vidéo sur les outils IA :
-
Tencent Launches HunyuanVideo-Avatar AI Video Generation Model
By
–
Tencent launched HunyuanVideo-Avatar, a new AI video model
— Rowan Cheung (@rowancheung) 2 juin 2025
It turns a single reference image and audio into a realistic short video with lifelike speech/singing
Suitable for domains like ecommerce and advertisingpic.twitter.com/QR3PGrR9N4Tencent launched HunyuanVideo-Avatar, a new AI video model It turns a single reference image and audio into a realistic short video with lifelike speech/singing Suitable for domains like ecommerce and advertising
-
Hume Launches EVI 3 AI with Advanced Voice Interaction Capabilities
By
–
Hume launched EVI 3, a new AI that can speak in any voice or personality via speech-to-speech interaction
— Rowan Cheung (@rowancheung) 2 juin 2025
It uses a voice-to-voice architecture and comes with a deeper understanding of tune, rhythm, timbre, and speaking stylepic.twitter.com/coHJKLV5TVHume launched EVI 3, a new AI that can speak in any voice or personality via speech-to-speech interaction It uses a voice-to-voice architecture and comes with a deeper understanding of tune, rhythm, timbre, and speaking style
-
Genspark AI Agent Downloads Media and Searches Web
By
–
ICYMI: https://t.co/4DRlGKYPRT has an Agent that can download images, PDFs and videos. It can perform a web search and reason across different tasks to make it work.
— 🚨 AI News | TestingCatalog (@testingcatalog) 1 juin 2025
AI Drive creates a folder for each task that you can browse after it is complete.
Works with X and YT 👀 pic.twitter.com/eFM4fLq2ZRICYMI: http://
genspark.ai has an Agent that can download images, PDFs and videos. It can perform a web search and reason across different tasks to make it work. AI Drive creates a folder for each task that you can browse after it is complete. Works with X and YT -
AI struggles with color in pastel scene generation
By
–
I've noticed it can sometimes struggle to add color to a pastel scene
-
Top AI Tools 2026: ChatGPT, Claude, Suno, Runway and More
By
–
ChatGPT pro, Claude Max, X premium (for Grok), Replit, Suno, Runway, Hedra, @augiestudio (video), @_glasp (bookmark), @DocDraftai (legal), @uselayers (marketing), @flymyai (media inference), @MageSpace_ (reusable image styles), @composio (tool use)….
-
Testing Autonomous Drone with Embedded Multimodal LLM
By
–
brb gonna test fly an autonomous drone with a small embedded multimodal llm
-
Kontext Chat: Edit Images Through Conversational Interface
By
–
Introducing Kontext Chat. Edit images by just chatting with them.
— Replicate (@replicate) 1 juin 2025
Kontext was crying out for a chat interface, so we made one.
Check it out: https://t.co/2upYgR0PnM pic.twitter.com/V6cXdBYfjqIntroducing Kontext Chat. Edit images by just chatting with them. Kontext was crying out for a chat interface, so we made one. Check it out: http://
kontext-chat.replicate.dev