Software companies are constantly trying to add more and more AI features to their platforms, and AI companies are constantly releasing new models and features.
Here are all the major AI updates we covered in the month of April.
Anthropic announces Integrations for Claude
Integrations allows users to connect apps and tools to Claude using the Model Context Protocol (MCP). Previously, Claude’s support for MCP was limited to Claude Desktop using local servers, but now Claude can work with remote MCP servers across the web and desktop apps.
There are currently 10 Integrations to choose from, including Atlassian’s Jira and Confluence, Zapier, Cloudflare, Intercom, Asana, Square, Sentry, PayPal, Linear, and Plaid, with more to follow. Developers can also create their own Integrations.
“When you connect your tools to Claude, it gains deep context about your work—understanding project histories, task statuses, and organizational knowledge—and can take actions across every surface. Claude becomes a more informed collaborator, helping you execute complex projects in one place with expert assistance at every step,” Anthropic wrote in a blog post.
Amazon Nova Premier now generally available
Nova Premier is Amazon’s most capable foundation model, designed to handle complex tasks and be a teacher for model distillation.
It can take in text, image, or video (no audio) inputs and has a context length of one million tokens.
“With its advanced capabilities, Nova Premier excels at complex tasks that require deep understanding of context, multistep planning, and precise execution across multiple tools and data sources,” Amazon wrote in a blog post.
JetBrains open sources its code completion LLM, Mellum
JetBrains has announced that its code completion LLM, Mellum, is now available on Hugging Face as an open source model.
According to the company, Mellum is a “focal model,” meaning that it was built purposely for a specific task, rather than trying to be good at everything. “It’s designed to do one thing really well: code completion,” Anton Semenkin, senior product manager at JetBrains, and Michelle Frost, AI advocate at JetBrains, wrote in a blog post.
Focal models tend to be cheaper to run than general larger models, which makes them more accessible to teams that don’t have the resources to be running large models.
Akka announces new deployment options to benefit AI agents
Customers will now be able to self-manage Akka nodes and Akka Platform regions. According to the company, these new options for deployment benefit developers who are building distributed systems, such as agentic AI systems.
“Agentic AI has become a priority with enterprises everywhere as a new model that has the potential to replace enterprise software as we understand it today,” said Tyler Jewell, CEO of Akka. “With today’s announcement, we’re making it easy for enterprises to build their distributed systems, including agentic AI deployments, without having to commit to Akka’s Platform. Now, enterprise teams can quickly build scalable systems locally and run them on any infrastructure they want.”
Ketryx launches AI agents for regulated industries
The validated agents maintain human-in-the-loop oversight so that development teams in regulated industries can utilize AI while still maintaining compliance and oversight.
The agents include a Complaint Agent to streamline complaint intake and flag critical signals, a Redundancy Detection Agent to eliminate duplicate items, a Test Coverage Analysis Agent that evaluates and improves requirement test coverage, an Anomaly Review Agent that analyzes anomalies and suggests improvements, and a Change Request Review Agent that analyzes product modifications, traceability, and potential risks.
Docker MCP Catalog to launch next month with 100+ verified MCP tools
Docker is introducing new MCP-related offerings to provide developers with tools for working with the Model Context Protocol (MCP).
Coming in May, Docker MCP Catalog will be a marketplace where developers can discover verified and curated MCP tools. The company partnered with several companies to build the catalog, including Stripe, Elastic, Heroku, Pulumi, Grafana Labs, Kong, Neo4j, New Relic, and Continue.dev.
The catalog contains over 100 tools, and each tool comes with publisher verification, versioned releases, and curated collections.
Solo.io launches Agent Gateway, Agent Mesh
Agent Gateway is an open source data plane that provides security, observability, and governance for both agent-to-agent and agent-to-tool communication. It supports popular interoperability protocols like Agent2Agent (A2A) and Model Context Protocol (MCP), and also integrates with agent frameworks like LangGraph, AutoGen, Agents SDK, kagent, and Claude Desktop.
Agent Mesh provides security, observability, discovery, and governance across all agent interactions, no matter where the agents are deployed. Key capabilities include multitenant across boundaries and controls, standard agent connectivity with A2A and MCP, automated collection and centralized reporting of agent telemetry, and a self-service agent developer portal to support discovery, configuration, observeability, and debugging tools.
AWS creates new benchmark for AI Coding Agents
SWE-PolyBench is a benchmark that evaluates the coding abilities of AI agents. It includes more than 2,000 curated issues in four different languages (Java, JavaScript, TypeScript, and Python), a stratified subset of 500 issues for rapid experimentation, a leaderboard with a rich set of metrics, and a variety of tasks, encompassing bug fixes, feature requests, and code refactoring.
The benchmark is publicly available and its dataset can be accessed on HuggingFace. There is also a paper about SWE-PolyBench on arxiv.
“This open approach invites the global developer community to build upon this work and advance the field of AI-assisted software engineering. As coding agents continue to evolve, benchmarks like SWE-PolyBench play a crucial role in ensuring they can meet the diverse needs of real-world software development across multiple programming languages and task types,” AWS wrote in a blog post.
OpenAI adds image generation model to API
OpenAI released its latest image generation model, gpt-image-1, in ChatGPT in March, and earlier this month, that model was added to the API. This addition will enable developers to add image generation capabilities into their own applications.
“The model’s versatility allows it to create images across diverse styles, faithfully follow custom guidelines, leverage world knowledge, and accurately render text—unlocking countless practical applications across multiple domains,” OpenAI wrote in a blog post.
NVIDIA NeMo microservices now available
NVIDIA NeMo microservices provide developers with a platform for creating and deploying AI workflows. Developers can use it to create agents that are enhanced with enterprise data, and can take user preferences into account.
Some of the microservices included in NVIDIA NeMo are:
- NeMo Customizer, which utilizes post-training techniques to accelerate fine-tuning
- NeMo Evaluator, which simplifies evaluating AI models on popular benchmarks
- NeMo Guardrails, which helps developers implement compliance and security safeguards
“The microservices have become generally available at a time when enterprises are building large-scale multi-agent systems, where hundreds of specialized agents — with distinct goals and workflows — collaborate to tackle complex tasks as digital teammates, working alongside employees to assist, augment and accelerate work across functions,” NVIDIA wrote in a blog post.
Zencoder acquires Machinet to further improve its AI coding agents
Zencoder, a company that provides an AI coding agent, has announced that it acquired another company in the AI coding agent business: Machinet.
According to Zencoder, this acquisition will solidify the company’s position in the AI coding assistant market and enable it to expand its multi-integration ecosystem into more development environments.
Machinet is a plugin for JetBrains IDEs, and while Zencoder already supported JetBrains, Machinet had even more specialized expertise in the ecosystem.
Machinet’s domain and marketplace presence will be transferred to Zencoder, and current Machinet customers will receive instructions on how to transition to Zencoder’s platform.
Vercacode adds new AI capabilities to its DAST offering
The latest capabilities are designed to enable organizations to respond to security threats more quickly. The new Enterprise Mode in DAST Essentials includes features like advanced crawling and auditing, AI-assisted auto-login to reduce authentication failures, Internal Scan Management (ISM), an intuitive interface, and real-time flaw reporting.
“DAST Enterprise Mode empowers security teams to work faster, smarter, and safer,” said Derek Maki, head of product at Veracode. “With real-time analysis in a unified platform, it eliminates the challenge of fragmented tools and enables mature, resilient risk management with centralized visibility and control.”
OpenAI reveals several new models and tools
The announcements from OpenAI include:
NVIDIA releases AI-Q Blueprint
AI-Q can help developers build agentic systems that use reasoning to “unlock knowledge in enterprise data.”
NVIDIA’s Blueprints are pre-defined and customizable workflows that developers can adapt, and the AI-Q Blueprint includes a reference for integrating with NVIDIA accelerated computing, partner storage platforms, and software and tools.
“AI-Q offers a powerful foundation for enterprises to build digital workforces that break down agentic silos and are capable of handling complex tasks with high accuracy and speed,” NVIDIA wrote in a blog post.
Anthropic announces Research and Google Workspace integration
Research is a new feature in Claude that allows it to search across internal work context and the web. It conducts multiple searches, each one building on the previous one to decide what to search next. It also explores several different angles of a question and works through them systematically.
Claude also now integrates with a few Google Workspace apps: Gmail, Calendar, and Google Docs. This will give it added personal context, such as emails or calendar commitments.
“Earlier this year, we introduced our vision for Claude as your collaborative partner that delivers hours of work in minutes. In line with this vision, we’re continuing to expand the context that Claude has access to, as well as the kinds of outputs you can generate with Claude to support your personal and professional work,” the company wrote in a post.
Veo 2 now in Gemini and Whisk
Veo 2 is a video model that can produce high-resolution, detailed videos from text prompts. Users can create video clips that are eight seconds long, 720p resolution, and in a 16:9 format.
Veo 2 in Gemini is available to Advanced subscribers and is available in Whisk for Google One AI Premium subscribers.
GitLab Duo with Amazon Q now available
With this integration, Amazon Q agents are embedded directly into GitLab’s DevSecOps platform. They can help with tasks like feature development, codebase modernization, vulnerability remediation, and code review optimization.
The solution is available through a bundle for GitLab Ultimate self-managed customers on AWS.
JetBrains announces a free tier for its AI tools
JetBrains has been adding several AI offerings to its portfolio over the past few years, including AI Assistant and its AI agent Junie.
Now, the company is announcing that all of those AI tools will be available under a single subscription.
As part of this announcement, the company said that it will offer a free tier that gives unlimited code completion and access to local AI models. Users on the free tier will be given a small cloud credit quota for using cloud-based AI assistance, including Junie. The free tier also comes with 30 days of AI Pro access.
Symbiotic Security launches AI tool for detecting and fixing vulnerabilities in code
Symbiotic Security is releasing a new tool that will enable automatic detection and remediation of vulnerabilities in code.
Embedded directly into a developer’s IDE, Symbiotic Security Version 1 utilizes an AI model that was trained on a “proprietary, security-specific, and verified dataset.”
In addition to detecting and remediating issues, it also features a built-in chatbot that developers can converse with to build their security skills. It provides training related to specific vulnerabilities, such as explaining remediation strategies and generating suggested fixes. The company likens this to spellcheck, but for vulnerabilities.
Moveworks creates AI Agent Marketplace
The new marketplace provides access to hundreds of pre-built AI agents that can be installed and deployed. As part of the announcement, the company also announced partnerships with Snowflake, Databricks, Stack Overflow, and Highspot, all of whom will be contributing their own agents to the AI Agent Marketplace.
The AI Agent Marketplace is a part of Moveworks Agent Studio, which also got several updates, such as the addition of Agentic Automation Engine, Agentic Reasoning Engine, and Plugin Workspace.
Google announces new and updated tools for building AI agents
The company announced the Agent Development Kit (ADK), an open-source framework that covers the end-to-end process of building and deploying agents and multi-agent systems.
Google also announced the Agent2Agent (A2A) protocol, an open protocol that allows AI agents to communicate with each other, exchange information in a secure way, and coordinate actions on top of enterprise applications. It was developed with support and contributions from over 50 other companies, including Atlassian, Box, MongoDB, Salesforce, ServiceNow.
In Agentspace, organizations will now be able to give their employees access to the platform’s unified enterprise search, analysis, and synthesis capabilities from within Chrome’s search box. Other new capabilities include a no-code Agent Designer for building custom agents, and access to two new agents built by Google: Deep Research and Idea Generation.
GitHub Copilot adds agent mode, MCP support in latest release
A new version of GitHub Copilot has begun rolling out to users, adding a number of new capabilities, such as the addition of agent mode to VS Code.
With agent mode, Copilot can iterate across an entire project, suggesting terminal commands, analyzing run-time errors, and more. “With simple prompts, agent mode takes Copilot beyond answering a question, instead completing all necessary subtasks across automatically identified or generated files to ensure your primary goal is achieved,” the company’s CEO Thomas Dohmke wrote in a blog post.
As an example, GitHub published a demo video in which a developer needs to update a website for runners that allows them to sort races by name, distance, and time. Upon receiving the request, Copilot analyzes the site to determine what needs to be changed, then begins by updating the backend and the UI, then generating unit tests for both, before handing it back over to the user to run those tests.
Microsoft updates Copilot to be a more personalized AI companion
According to Microsoft, the latest updates are designed to bring it from just an “AI companion” to “your AI companion.” It will now be able to remember what you talk about to learn your likes and dislikes, as well as details about your life, such as your dog’s name or work projects you’re assigned to.
Other new capabilities added to Copilot include:
- Deep Research, for conducting multi-step research tasks
- Actions, for completing tasks on your behalf, such as booking event tickets or making dinner reservations
- Pages, which combines notes, content, and other research into a single canvas
- Vision, which lets Copilot see what you see and respond in real time
“Copilot will understand you in the context of your life, and show up, on your terms, in the right way at the right time. This is far richer, more dynamic, supportive and emergent than any software we’ve seen before. It’s a new kind of relationship with technology, a new era,” Microsoft wrote in a blog post.
OpenAI expands ChatGPT’s memory capabilities
In an effort to get ChatGPT to know more about you, OpenAI has announced that it can now reference all past chats that have occurred. This update “points at something we are excited about: ai systems that get to know you over your life, and become extremely useful and personalized,” Sam Altman said on X.
This feature began rolling out earlier this month for Pro users, and Plus users will soon get it as well. Users will be able to opt out of this, and can always use the temporary chat feature to have a conversation that won’t be saved to ChatGPT’s memory.
Amazon Nova Sonic is announced
This is the latest model to be added to the Nova family, and it is a speech-to-speech model for conversational AI.
According to Amazon, typically, speech understanding and generation are covered by two separate models, but Nova Sonic unifies both into one model. “The result is an adaptive speech response that dynamically adjusts its delivery based on prosody, such as pace and timbre, of input speech,” Amazon wrote in a blog post.
Gemini 2.5 Pro now in public preview
Due to the adoption of Gemini 2.5 Pro and feedback from users, Google has decided to move it to public preview in the Gemini API in Google AI Studio. It will also be accessible through Vertex AI shortly, Google said.
The pricing for Gemini 2.5 Pro is $1.25/1 million tokens for text, image, audio, and video inputs in a 200k context window or less, and $2.50/1 million tokens text greater than a 200k context window. The output pricing is $10/1 million tokens for context windows under 200k and $15.00/1 million tokens for context windows greater than that. The experimental version will remain free, but will have lower rate limits.
Meta releases first Llama 4 models
The announcement covered three models that are part of the Llama 4 family:
- Llama 4 Behemoth (in preview), a teacher model for distillation that offers 288B active parameter, 16 experts, and 2T total parameters
- Llama 4 Maverick, a native multimodal model with 1M context length, offering 17B active parameters, 128 experts, and 400B total parameters
- Llama 4 Scout, optimized for inference with a 10M context length, that has 17B active parameters, 16 experts, and 109B total parameters
Cloudflare announces remote MCP server to reduce barriers to creating AI agents
Cloudflare is making it easier for developers to create AI agents with several new updates that were announced earlier in the month.
First, the company announced a remote Model Context Protocol (MCP) server. MCP is an open standard that allows AI agents to interact with external services, which is important in enabling them to complete tasks on a user’s behalf.
According to Cloudflare, MCP has thus far been limited to running locally, which has prevented its mainstream adoption. Now, users will be able to build and deploy remote MCP servers through Cloudflare, which will allow agents to connect and interact with services without needing to rely on a locally hosted server.
Progress packs Sitefinity 15.3 with AI updates
The latest version of the content management system offers AI capabilities for users like summarizing content, improving writing, personalizing messages, and generating tag suggestions. Users will also be able to search for images using natural language descriptions.
Other AI capabilities include AI translation, content optimization, and the ability to export marketing data in the Parquet format.
“AI is revolutionizing what is possible in creating digital content and experiences, and any organizations not using these capabilities will be at a significant competitive disadvantage,” said Loren Jarrett, EVP & GM of digital experience of Progress. “Sitefinity is at the forefront of innovation, empowering marketers, developers and content teams with the most advanced tools so they can deliver digital experiences that will enable them to compete and win in an AI-driven world.”
Anthropic announces Claude for Education
This new offering includes a “Learning mode” that encourages students to work through problems alongside Claude, rather than having Claude come up with answers entirely on its own.
For example, Claude might respond saying “What evidence supports your conclusion?” to make students better understand the work they are doing.
It will also emphasize core concepts for students, and provides templates for how to structure research papers, study guides, and outlines.
Amazon creates new website for exploring Amazon Nova foundation models
nova.amazon.com provides information on each of the company’s Amazon Nova foundation models, such as Nova Micro, Nova Pro, and Nova Canvas. Each model has its own page that includes more information about the model, benchmarks, a demo, and resources on how to get started.
In addition to launching this new website, the company also announced Amazon Nova Act, which is a new model that can perform actions in a web browser. Amazon released a research preview of the SDK for the model so that developers can begin to experiment with it.
“Nova.amazon.com puts the power of Amazon’s frontier intelligence into the hands of every developer and tech enthusiast, making it easier than ever to explore the capabilities of Amazon Nova,” said Rohit Prasad, SVP of Amazon Artificial General Intelligence. “We’ve created this experience to inspire builders, so that they can quickly test their ideas with Nova models, and then implement them at scale in Amazon Bedrock. It is an exciting step forward for rapid exploration with AI, including bleeding-edge capabilities such as the Nova Act SDK for building agents that take actions on the web. We’re excited to see what they build and to hear their useful feedback.”
Solo.io launches MCP Gateway to tackle AI agent sprawl
Solo.io has announced the launch of its MCP Gateway, a Model Context Protocol gateway for the cloud native API gateway kgateway (previously called Gloo).
MCP is an open protocol developed by Anthropic that provides a standard for how applications connect data sources and tools to LLMs. According to Solo.io, as MCP adoption continues to grow, AI clients and agents are experiencing tool sprawl and facing difficulty in managing discovery, connectivity, and security when integrating with many different MCP-compatible tools.
MCP Gateway attempts to solve this problem by enabling developers to discover, secure, and federate multiple MCP tools and tool services into a virtualized MCP server. This serves as a single access point for developers, regardless of how many tools are in their AI agent ecosystem.
Kong AI Gateway updated with features to reduce LLM hallucination and protect sensitive personal data
Kong has announced updates to its AI Gateway, a platform for governance and security of LLMs and other AI resources.
One of the new features in AI Gateway 3.10 is a RAG Injector to reduce LLM hallucinations by automatically querying the vector database and inserting relevant data to ensure the LLM is augmenting the results with known knowledge sources, the company explained.
This improves security as well by putting the vector database behind the Kong AI Gateway, and also improves developer productivity by allowing them to focus on things other than attempting to reduce hallucinations.
Env0 announces Cloud Analyst
Cloud Analyst is an AI agent that provides insights into large organizations’ entire cloud infrastructure. Users can ask questions about infrastructure state, trends, and key metrics; explore historical data, filter by project or environment, and uncover potential opportunities for optimization; and create interactive dashboards.
It will be available as part of env0’s platform starting in Q2 of this year, and currently has a waitlist.
CodeSignal adds AI skills assessments
The assessments are designed to measure AI skills in the workplace. They include an AI Literacy Assessment, a Prompt Engineering Assessment, and an AI Researcher Assessment.
“More than 40% of workers will need to reskill in the next three years due to AI and automation, according to the World Economic Forum,” said Tigran Sloyan, CEO and co-founder of CodeSignal. “The AI Collection gives organizations a precise and scalable way to assess and develop those capabilities, whether they’re hiring for new roles or investing in the growth of their existing teams.”
Zencoder releases new AI coding and unit testing agents
The new agents integrate with over 20 different development tools, including VS Code, JetBrains IDEs, JIRA, GitHub, GitLab, and Sentry.
They are capable of complex, multi-file refactoring and merges, and can automatically test and refine outputs.
“We are on a journey to turn everyone into a 10x engineer and unlock their full potential and creativity,” said Andrew Filev, CEO and founder of Zencoder. “The agents we release today are a big step forward; they are a force multiplier that can handle routine coding, letting you focus on innovation and more complex aspects of software engineering.”