GitLab Transcend - Intelligent orchestration
Mini tools
AI Tools released in
From Presentations to Ai girlfriend to Personal assistant, browse the complete list of AI Tools released in the last 7 days. These tools are ready to enhance your everyday tasks with the latest technology.
-
Mikael Jones🙏 2 karmaJun 10, 2026@TranscripToolBeen using TranscriptTool whenever I need to quickly turn short videos into text. I like that it keeps things simple and makes it easier to review claims or pull quotes without having to watch clips repeatedly. Handy tool for anyone dealing with video content regularly. -
Open -
New Features & Enhancements: • AI Chatroom: Introduced a revolutionary multi-AI conversational environment. Users can now invite multiple AI models into a single chat session, directing messages to specific AIs for their input. This enables complex collaborative tasks, comparative analysis, or simulated debates between different models, with each AI having access to the full conversation history. Responses are manually triggered, providing users with precise control over the interaction flow. • AI Coder: Launched a dedicated, AI-powered coding workspace. This feature allows users to leverage AI models for code generation, debugging, and iterative development. Users can provide prompts to build new projects, import existing code files, and receive AI-generated code suggestions and modifications, complete with diffs to track changes. The Coder environment is designed to accelerate software development workflows by integrating powerful AI capabilities directly into the coding process. • Improved Model Browser Filters: Enhanced filtering options in the Model Browser for more granular discovery of AI models based on advanced capabilities and pricing tiers. • Performance Optimizations: General improvements across the platform for faster load times and a smoother user experience. Bug Fixes: • Resolved minor UI inconsistencies in the chat interface. • Addressed an issue causing occasional delays in model response times.
-
Claude — v5.0Claude Fable 5 State-of-the-art on Cognition's FrontierCode eval, scoring highest among frontier models even at medium effort. More token-efficient than prior Claude models. Stripe reported a codebase-wide migration on a 50M-line Ruby codebase done in a day, versus an estimated two-plus months by hand. Highest score of any model on Hebbia's Finance Benchmark (senior-level reasoning), with major gains in document reasoning, chart and table interpretation, and problem solving. IMC reported near-across-the-board top results on trading-analysis evals (factual lookup, conceptual reasoning, root-cause analysis, expected-value analysis). New state-of-the-art for vision tasks. Extracts precise numbers from scientific figures and can rebuild a web app's source code from screenshots alone. Needs less scaffolding: beat Pokémon FireRed with a minimal vision-only harness, where earlier models needed complex helper harnesses. Stays focused across millions of tokens on long-running tasks and improves its outputs using its own notes. With persistent file-based memory in Slay the Spire, performance improved 3x more than Opus 4.8, and it reached the final act 3x more often. Works autonomously for longer than any prior Claude model
-
Create Stunning PPTs in Seconds with the AI PPT MakerOpenMeet Maya — chat your way to a finished deck. Describe what you want, Maya builds the whole presentation, then edit any slide (or the whole thing) in plain language. Plus a wave of new ways to turn anything into slides. New in 2.0: • Maya — conversational AI that generates AND edits your full deck in chat • AI Pitch Deck Generator — describe your startup, get an investor-ready deck • MagicChart — turn data into interactive AI charts • AI Image Generator — create custom slide visuals from a prompt Turn anything into slides: • Audio & Video to PPT (incl. YouTube — auto-transcribed) • Image to PPT (screenshots, photos, scans) • Excel & Word to PPT • URL to PPT — any web page into a deck • Text to PPT — paste notes, get slides • PDF tools suite (PDF→PPT, PDF→JPG, and more) • Import from Gamma & Tome More: • Translate whole decks into any language (layout intact) • Remix any template with AI • Redesigned editor with more layouts • Export anywhere — PDF, JPG, MP4, or text • MagicSlides mobile app now on iOS & Android • Faster email-OTP signup + free trial -
Open -
More than a vibe-coding tool, Whacka helps users build, launch, run, and maintain their own apps, with updated resources and a smoother end-to-end experience in the new version.
-
-
It's great to have free tools of such high value
-
Honestly, MySpec saved us from a massive rewrite. We used to just start coding and hope the architecture would figure itself out (spoiler: it didn't, it was pure spaghetti). Using this tool completely changed how we plan. It forces you to get crystal-clear on the specs and scalability before you lay down a single line of code. If you’re tired of 3 AM tech-debt panic attacks, just try it. Seriously.
-
Really impressed with audien.to so far. I dropped in a meeting recording and got a clean transcript plus useful notes in just a few minutes, without having to tweak much. The speaker tags and filler cleanup make it way easier to share with the team.
-
Really impressed with Junior.so so far. Having an AI teammate inside Slack that can handle CRM updates and follow-ups without me constantly prompting it feels genuinely useful, especially for small teams juggling sales and ops. Setup was pretty smooth, and the daily reports are a nice bonus.
-
AudioScribe has been really helpful for turning meeting recordings into clean transcripts without much hassle. I especially like the AI summaries and speaker identification, since they make it much easier to review long conversations and pull out the key points. Solid tool overall, and definitely worth a 5-star rating.
-
-
AI agent with infrastructure knowledge for incidents.Open -
Notis is the AI intern one message away from your entire tool stack. Dictate ideas, delegate the busywork, and let Notis update your workflows across WhatsApp, Telegram, iMessage, Slack, and email. In this week's release: - Improved account safety by rejecting inbound messages from unlinked senders instead of creating accounts automatically. - Added one-time email verification links for the email channel. - Team billing now pool team usage credits together instead of using individual credits. - Improved tool and integration reliability, especially around Composio auth links, outage handling, and dynamic tool discovery. - Added a transcribe tool and upgraded web scraping to Firecrawl. - Disabled the SMS channel to stop fraud abuse (RCS and iMessage still works of course). - Added a billing overview on the user setting menu.
-
Let your AI agents run your Meta/Google/TikTok ads for you! Spy competitors, generate creatives, launch ads, and manage everything from your AI agent directly! Since v1.1.0: - TikTok Ads — create campaigns, upload media, and manage TikTok ads via MCP and CLI. - Reddit Ads — create and manage Reddit campaigns via MCP (beta). - Google Performance Max — build and publish PMAX campaigns with Merchant Center support (beta). - Studio → Meta publishing — generate ads in Studio and publish them straight to Meta, with per-placement media variants. - Workspaces — manage your team under a single workspace with shared access to all projects. - Multi-project MCP — connect all your projects with a single MCP connection. No more one-connector-per-project. - Multiple ad accounts per project — connect several Meta, Google, TikTok, or Reddit accounts to the same project. - ChatGPT support — connect AdKit to ChatGPT as an AI agent alongside Claude, Cursor, and Codex. - Advanced Platform Access — direct API access to Meta, Google, and TikTok for features not yet built into AdKit. - Self-serve API keys — create API keys from the dashboard for agents that don't support remote MCP.
-
Tweet Ideas, Reddit Voting and more ✨ New: Tweet Ideas Pounce now distills your best replies from a session into ready-to-post tweet ideas — edit them, post straight to X, or copy them for later. 🔼 Reddit Voting Upvote and downvote Reddit posts right from your queue — engage without ever leaving Pounce. ⚡ Faster, Smoother Sessions 1. Keep replying while a session is paused — no need to resume first (paid plans). 2. Connecting a platform now drops you straight into authorization instead of stopping at an extra screen first. 3. New campaigns from an X setup name themselves automatically based on what you're promoting, instead of landing as another "My Campaign".
-
Updated plans: - Basic plan available now starting at $39/mth. New features: - You can call Edith on the phone - Experimental: skills & mini apps - Telegram first for messaging
-
1. Reviewable, Accuracy-First for Video Courses 2. ALL in One For Accurate Course Videos 3. The Accuracy First For Educational videos 4. 150+ Knowledge Visualization Box, 3D/SVG/Video/Image/Chart... 5. BrandKit && Team, Google VEO
-
Been experimenting with KreoGrid for content ideas, and I like how it turns trends into something you can actually publish. Instead of spending hours researching what's working, it helps speed up the process and makes campaign planning feel much more streamlined.
-
OpenBeen using Arobis AI Visibility Checker to get a better sense of how content appears across AI-driven search and recommendation systems. The reports are easy to understand and make it much simpler to spot areas that could be improved. Helpful tool if you're trying to understand and grow your AI visibility. -
Been using Knowcast to keep up with new ideas and information, and I like how easy it is to discover and revisit useful content. The experience feels straightforward and helps cut through a lot of the noise, making it easier to stay informed without getting overwhelmed.
-
Been playing around with Edge Arena and I like the concept of comparing different AI approaches before committing to one. It’s useful for turning ideas into more actionable plans, and the focus on execution rather than just brainstorming makes it stand out from a lot of other AI tools.
-
Reve Image — v2Planning and rendering are separated so layouts can be inspected, controlled, and edited before final image generation. Code-based image representation makes each element more addressable for direct manipulation and agentic editing. Native 4K x 4K rendering enables 16MP outputs without relying on a separate upscaling step. Longer iterative editing stays more stable with less reference-image degradation and no artifact buildup when editing from code. Stronger composition, text placement, environmental typography, lighting, and spatial detail support more production-ready design workflows.
-
Crucible 2.0 is live. A workspace for business decisions, not a chat box. New: Decision workspace (frame → debate → Decision Brief), PERSPECTIVE.md personas (including your own voice from CV or text), Documents, Knowledge graph, Memory Search, Persistent memory (Pro), MCP Connect to Claude & Cursor. Pricing: Free (1 debate/30 days), Starter $19/mo, Pro $45/mo. crucible.roundtablelabs.ai
-
Now with full multilingual support. Use Wiki 2066 in English, German, or any language your browser displays, with a smooth, stable experience from start to finish. Generate a 2066 biography of anyone: yourself, someone you know, a public figure, or a fictional character. Every wiki is built on the ninety-year timeline from The Singularity Roadmap 2026–2116, the book's detailed forecast of AI acceleration, longevity medicine, and technological change is the backbone of every biography the app writes. A BYOK app: bring your own Anthropic and xAI keys and pay the providers directly at published rates, with no markup. Your keys, profile, and generated wiki stay entirely in your browser. No install required.
-
-
The Ultimate All-in-One AI Podcast Studio! 🎙️🎞️ PodcastorAI is by far the most complete podcast tool I’ve found on Toolify. It seamlessly bridges the gap between Audio Podcast and Video Podcast. You can literally go from a raw prompt or document to a fully polished, multi-host video show ready for YouTube and Spotify. The interface is intuitive, the voice options are highly expressive, and the visual layouts (like Split Screen & Two-Shot) look amazingly professional. Incredible efficiency booster for any content team!
-
AI-powered restaurant location analysis before you lease.Open -
Linkeddit now watches Reddit for you around the clock. Where the lead gen pipeline runs a one-time search, Reddit Monitors run on a schedule and surface new matching posts automatically — so buying-intent threads, brand mentions, and competitor complaints show up without you re-running anything. How it works: - Set up to 10 keywords and, optionally, up to 10 subreddit filters. Leave subreddits empty to scan all of Reddit. - Choose a frequency (daily, weekly, or monthly), a Reddit sort (new, relevance, hot, top, comments), a time window, and a result limit per run. - Give each monitor a knowledge base — your product, pricing, voice, and "do not" rules — so every AI reply is grounded in what you actually offer. - Add optional writing-style instructions to match your voice exactly. AI reply suggestions: click any matched post and Linkeddit drafts a single, ready-to-send Reddit comment grounded in the original poster's exact words and your knowledge base. Replies use a Reddit-native voice — no AI tells, no emojis, proper Reddit markdown — and only mention your product when it is genuinely a useful answer. Run on autopilot or trigger a monitor manually any time. Status tracking shows what is queued, running, or paused. Learn more: linkeddit.com/reddit-monitoring
-
Start from real examples. Turn one good clip into a full video.OpenAdd Story mode: users can create continues video clips without losing characters or key subjects. -
-
Open -
Turn simple prompts into viral videos, stunning images, and realistic audio.Open -
Smarter AI Music Generation, Better Covers, More Realistic AI Music Videos AirMusic v2.4 delivers significant improvements across music creation, cover generation, and AI video synthesis — making the entire AI music to video workflow more reliable, realistic, and rewarding. This release focuses on quality upgrades, new models, and a recommend‑and‑earn credits feature to thank our community. 🎶 AI Music Generator — Gemini Lyria 3 Integration We’ve added the Gemini Lyria 3 model to the AI Music Generator. - Significantly better adherence to prompt instructions - More coherent melodies, tighter structure, and higher musicality - Improved output for mood, genre, and style specificity Whether you want lo‑fi beats, pop hooks, or cinematic themes, the new model helps you generate tracks that match your prompt more faithfully. 🎤 AI Cover & Music Extender — Higher Success Rates AI Cover and Music Extender have been optimized to improve overall reliability. - Generation success rate increased by 78% - Fewer errors, fewer retries - Covers sound closer to target genre and vocal style - Extended tracks maintain musical coherence and rhythm These improvements make remixing, extending, and reinvigorating music far less error‑prone. 💎 Recommend & Earn — Get Credits for Referrals Introducing our new recommendation rewards system: - Successfully refer a new paying user - You earn 200 credits Credits can be used for music generation, covers, videos, and more — making it easier to create high‑quality content without paying every time. 📹 AI Music Video Generator — Higher Fidelity Visuals AirMusic’s AI Music Video Generator now supports HappyHorse as a reference‑to‑video model. - Significantly improved visual fidelity when using image references - Characters, backgrounds, and details more consistent and realistic - Better handling of complex scenes like performances and environments This upgrade especially benefits creators making storytelling, singing, or performance‑focused music videos. 🧑🎤 Hunyuan Avatar — More Realistic Lip‑Sync We’ve added the Hunyuan Avatar model to singing mode. - Character mouth movement now better matches vocals - Smoother, more believable lip‑sync performance - Improved facial animation continuity This makes AI music videos feel more natural and expressive. 🧠 Quality‑First Updates Based on user feedback and data analysis, AirMusic has removed a set of lower‑quality models in: - Music generation - Image generation - Video generation This cleanup improves overall input quality, reduces noise and artifacts, and ensures you get higher‑quality output across the platform. 📈 Summary — What You Get in v2.4 - More accurate AI music generation with Gemini Lyria 3 - Higher success rates for AI Cover and Music Extender - Earn credits by referring paid users (+200 credits each) - Better visuals in AI music videos with HappyHorse - More realistic lip‑sync with Hunyuan Avatar - Cleaner model lineup for higher output quality With AirMusic v2.4, you’re getting a smarter, faster, and more reliable AI music creation and AI music video generator experience — ideal for creators who want professional‑level results with less friction.
-
-
I run my own business on this platform and its very effective, continuously improving, and focused on the human as the orchestrator.
-
-
Most sales teams lose meetings to scheduling friction. This release removes it. Reply now handles everything between "I'm interested" and "it's in the calendar" — automatically, without the back and forth that kills momentum. 📆 Book meetings directly from sequences with Calendly. Connect Calendly to Reply and drop your meeting link into any sequence. When a prospect is ready to talk, they book right from your message. — Meetings land instantly when interest is highest — Sequences stop automatically after a booking — Status updates sync when someone schedules or cancels — Full visibility into which outreach drives meetings No coordination required. Just booked calls. 📆 Jason AI handles scheduling end to end. Jason can now check your availability, book meetings, and manage cancellations on your behalf. All activity syncs directly to the prospect's card so nothing slips through. Three steps: Connect. Enable. Automate. 📆 Custom calendars for smarter send timing. Build a calendar that reflects your actual schedule — public holidays, vacation days, conferences, time off. Sequences skip those days automatically, so your outreach never lands on a day you're unavailable. Ready-made calendars for Australia, Canada, France, Germany, the UK, and the US are still available. Custom calendars give you full control. Fewer missed signals. Smarter timing. More meetings, with less effort. Start your free trial and see what Reply can do now.
-
docAnalyzer has been rebuilt around a bigger idea: you leave with files, not just an answer. When we first launched, docAnalyzer was a way to chat with your documents. Now the answer is the start. Ask across hundreds of sources and the workspace builds what you need out of that work: a report, a spreadsheet, a slide deck you can download and reuse. What’s new: - Chat to deliverable: compose answers into PDF, Word, Excel, slides, charts, and diagrams, and reuse any output in a later turn. - A workspace, not a chat box: outputs build across turns instead of resetting. - Workflows are now agentic: they run multiple rounds of search and read across your whole library instead of a single pass, with new ones added like SEO metadata and a Humanizer. - Three chat modes, one engine: a new Ask docAnalyzer mode plans and runs work across your whole workspace, alongside Focus chat (deep analysis on a chosen dataset) and Co-work chat (shape a draft in a canvas). - A rebuilt interface, with visibility into what the agent is doing while it works. If you knew docAnalyzer as “chat with your documents,” it is worth another look. The Community plan is free, and your first session is ready in about 30 seconds.
-
This is a really good site if you're trying to find a an easy way to find good products without wasting any time. this is the best
-
Save hours by auto-replying to YouTube comments with context-aware AI.Open -
We launched a rich set of photo merging tools: Add Person to Photo Put Two People Together AI Couple Photo Maker AI Family Portrait Generator Side by Side Photo AI Face Merger Including practical tools that can enhance the effect of image synthesis: Restore Old Photos Remove Background AI Photo Enhancer AI Headshot Generator and lots more...
-
Uramaki Studio AI just got a major upgrade. Create sharper, on-brand marketing assets faster — now with better logo handling, enhanced brand controls, TikTok image carousel support, editable fonts, colors and copy directly in preview mode, plus faster generation and project duplication. Free users can now create up to 5 projects with 5 slides each. Built for founders, growing teams and marketers who need high-quality carousels without slowing down execution. Try Uramaki Studio AI for free and scale your content workflow!
-
OpenMost AI tools start from zero. You paste in context, re-explain your work, and hope the output is relevant. Every. Single. Time. remio is different. It automatically captures everything you work on — web pages you browse, documents you read, meetings you record, emails you send — and turns it all into a living, private knowledge base. No manual tagging. No copy-pasting. No setup. Then, when you need to get something done, remio already knows your context. Write a doc, draft an email, generate a report, create slides — just tell it what you need. It pulls from your actual work history to produce output that's genuinely relevant to you. What makes remio different: 🧠 Passive, automatic context capture — works silently in the background 🔒 Local-first & privacy-focused — your data stays on your device ⚡ BYOK support — bring your own API key from OpenAI, Google, DeepSeek and more 🖥️ Local model support via Ollama 📎 Connects to Gmail, Google Docs, local files, YouTube, and more 🛠️ Build reusable aApp workflows powered by your own knowledge This is not another note-taking app. This is your personal agent — one that already knows your world. From knowledge to getting things done. 👉 Try remio today. -
Ideogram — v4Open weights enable downloading, fine-tuning, and running generations on your own hardware. Multilingual text rendering supports design work beyond English-only typography. Bounding-box layout control gives more precise placement for text, logos, and visual elements. Realistic 2K image output improves production use for ads, branding, apparel, photography, and social content. Native editable outputs are planned to return alpha channels and editable text layers directly from inference.
-
Turn any song into scroll-stopping videos in minutes with AI-directed scripts andOpen🎬 BeatMV v2.4 — Product Update Introducing creative styles, concept revision, a referral system, and a streamlined model lineup. ✨ New Features 🪩 Disco Mode & Y2K Aesthetic Two new style presets that instantly set the visual direction of your music video — from retro disco vibes with mirror balls and neon lighting, to nostalgic Y2K aesthetics. 💬 Concept Revision Refine your AI-generated video concepts with natural language feedback. Tell the AI what to change and get updated concepts without starting from scratch. (3 credits per revision) 👥 Invite Friends & Referral System A new "Invite Friends" badge in the header with native Web Share support on mobile. Includes anti-abuse protections and admin monitoring dashboard. 🎬 Sample Video Showcase Six curated MV Director sample videos now showcase the platform's capabilities, with the top 3 upgraded to uncompressed HQ R2 URLs. 🔧 Improvements & Fixes ⚡ Gemini 2.5 Flash-Lite Swapped Gemini 2.0-flash to 2.5-flash-lite across all 9 AI routes for faster responses and lower latency. 🧹 Model Cleanup Removed underperforming models (Seedance 1.0, SeedDream 4, Wan 2.7) and renamed the Nano series for clarity. 🖱️ Character Hover Preview Hovering over a character in the selector now shows a larger preview image, making it easier to pick the right character. 🖼️ Batch Image Model Fix Batch image generation now uses nano-banana-2 instead of seedream-v4 for better quality and reliability. ---
122 total records
