AI.info
AI Tools Updates — Page 180
What changed in the AI tools this week — dated, from the makers’ own changelogs.
En anglais
- Synthflow AI Voice Agent Builder — ElevenLabs V3, Synthflow TTS, MCP Servers, Outbound Number Pooling & more
ElevenLabs V3, Synthflow TTS, MCP server connections, and outbound number pooling.
- Leena AI Work Assistant — Product Newsletter — May 2026
- HeyGen — HyperFrames Render API — Decomposed Resolution + Aspect Ratio
resolution is now a tier; aspect_ratio is a separate field.
- Arcee Orchestra — Trinity Models Move to OpenMDW-1.1 License from Apache 2.0
Trinity Models Move to OpenMDW-1.1 License from Apache 2.0.
- Ambience — Ambience Healthcare Expands Chart-Aware Intelligence to Full Inpatient Workflow
Delivers clinical context across the full hospital stay, from admission through discharge.
- Arcee Orchestra — Trinity is moving to OpenMDW-1.1
Trinity is moving to OpenMDW-1.1.
- Relevance AI Workforce — Timeline view added to Tasks page
The Approvals, Escalated, and Errors tabs on the Tasks page now show a timeline chart of task counts per day.
- Jan — Jan v0.8.1: Anthropic-Compatible Custom Providers, Per-Message Errors & llama.cpp Settings Overhaul
Adds Anthropic-compatible custom providers, OS-native TLS trust, a sampler popover, per-thread and per-message errors, and llama.cpp settings.
- MLflow — MLflow 3.13.0 Highlights: Role-Based Access Control, Trace Archival, Coding Agents, and Hermes Agent Support
Define roles and assign them to users; archive traces; onboard coding agents; choose MLflow Assistant engines; deploy with Helm; trace Hermes Agent; set span log levels.
- Appsmith — Release v2.1 🌈
Replaced Intercom with Pylon across the platform.
- Ludwig — v0.17.5
GPU Docker images now install CUDA-enabled PyTorch wheels.
- Gamma — What's New in Gamma | May 29
Claude and ChatGPT connectors follow instructions more reliably; you can delete decks through the public API; five export improvements are live.
- Squibler — AI Story Generation Upgrade
Story generation flows now run on our provisioned AI model; fixed documents failing to load due to a missing snapshot file.
- LALAL.AI Stem Splitter — Fixed PayPal upgrades so the license stays active if an upgrade doesn't finish.
- Nayya Use — Cost Engine: How We Solved Healthcare's Biggest Blind Spot
Here's how we built the infrastructure to change that.
- Squibler — May 28th, 2026
Story generation now runs on a provisioned AI model; fixed document loading, blog links, and monthly price display.
- Squibler — Improvements
Story generation flows now run on our latest provisioned AI model, delivering faster and more consistent results when writing or expanding your story.
- LALAL.AI Stem Splitter — 2026-05-28
Fixed PayPal upgrades so the license stays active if an upgrade doesn't finish.
- Solve — AI Studio: Where Your AI Agents Learn, Improve, and Run Themselves
AI agents learn, improve, and run themselves
- Poolside Assistant — Poolside on Dell: an efficient path for frontier AI inside your boundary
Poolside running on Dell gives enterprises another path: AI that runs inside their own boundary, closer to their data.
- Bland — Evals
Evals run LLM judges against call history to grade quality dimensions; agents use Transcript or Audio mode.
- Microsoft Copilot — Introducing Microsoft 365 Business with Copilot: The new standard for small business
Microsoft is introducing new Microsoft 365 SKUs with Copilot built in for small businesses.
- Trinka AI — Trinka AI Launches Moodle Plugin, Offering Writing Support and Academic Integrity Within University’s Existing LMS
Trinka AI announced the launch of its Moodle Integration Plugin that enables universities to deploy Trinka's writing support and academic integrity solution directly…
- Typefully — Better, Smoother Realistic Previews
Updated previews to open in the editor sidebar, scale to fit mobile screens, and reflect changes to X and Threads previews.
- Pika — Get Your Product Out There
Four specialized Pika MCP skills to build a brand, generate app store screens, a sizzle video, and a founder video.
- Brand24 — New AI Visibility View: Sources
Brand24 added a Sources tab in AI Visibility to show citation sources, source comparisons, source types, and AI models.
- Flux — FLUX VTO: Virtual Try-On at scale
Virtual Try-On at scale.
- LALAL.AI Stem Splitter — Fixes:
Fixed PayPal upgrades so the license stays active if an upgrade doesn't finish.
- Vizcom — Your asset library just became a part of every generation
Two new asset types, Prompts and Collections, and a new way to use them with @ referencing in Modify. Save what works, reference it anywhere.
- Biela.dev — v3.0.24
Claude Opus 4.7; OpenRouter Catalog API and filters; GitHub branches, PRs, commits history; Mobile App Preview (Expo).
- Biela.dev — v3.0.23
Claude Opus 4.6, Gemini Flash, GLM 5; phone verification flow; OnBiela Subdomains; billing cards and yearly discount display.
- Biela.dev — v3.0.22
Clarifying Questions Tool; Tool Approval System; import projects from Bolt, Lovable, Replit; monthly free prompts and low-balance notifications.
- Mobileye SuperVision — Why learning from data gets harder in the tail
- Mobileye SuperVision — Diagnosing the long tail: how Mobileye turns edge cases into targeted training
Mobileye turns edge cases into targeted training.
- Hopper — WestJet Adds HTS Cancellation Offering
Lets travelers turn a nonrefundable fare into a refundable fare for an extra fee.
- Hopper Cloud (HTS) — Cancel For Any Reason powered by HTS - Hopper Technology Solutions arrives at WestJet
Cancel For Any Reason powered by HTS arrives at WestJet.
- InvokeAI — InvokeAI 6.13.0
Adds support for Qwen Image, Qwen Image Edit, Anima, and externally-hosted models from OpenAI, Gemini, BytePlus, and Alibaba Cloud.
- Wiz Defend — Evidence at the Moment of Attack. Answers at AI Speed.
Wiz Sensor Forensics is now generally available, automatically capturing forensic artifacts at the moment of detection and using AI to accelerate investigation.
- Read AI — Read AI is Now a Claude Connector
Meetings, action items, and customer conversations are building blocks inside Claude, with three new skills.
- Retell AI — Agent Versioning 2.0, Granular Multilingual, Custom Voicemail, Voice Orb, and more
Create multiple drafts, track who changed what, select languages, and customize how each agent detects voicemail.