xAI Updates & Release Notes

Use your SuperGrok or X Premium subscription inside OpenCode.

After connecting your Grok account, start coding with Grok Build, the same model that powers xAI's terminal-based coding agent.

Setup

Install OpenCode:

curl -fsSL https://opencode.ai/install | bash

Launch OpenCode:

opencode

Inside OpenCode, run /connect and pick xAI:

/connect

Pick the sign-in method that fits your setup — both use your Grok subscription:

• xAI Grok OAuth (SuperGrok Subscription) — opens your browser to sign in.
• xAI Grok OAuth (Headless / Remote / VPS) — prints a code and URL for SSH or remote hosts.

More open-source agents and integrations are coming soon.

Original source

Persistent expertise for Grok. Generate documents, decks, and spreadsheets. Automate workflows. Build and share your own skills.

Today, we're launching Skills. Teach Grok something once and it remembers across every conversation. No more repeating your preferences, formatting rules, or workflow steps.

Ready Out of the Box

Every Grok account ships with built-in skills from xAI, with no setup required. If you want to customize one, just override it with your own version. Yours always takes priority.

Word Documents

Create and edit .docx with full formatting

Presentations

Build slide decks from scratch

Spreadsheets

Generate formatted spreadsheets and analyze data

PDFs

Create, merge, split, and extract from PDFs

Skill Creator

Build new custom skills through conversation

Instant Office Documents

Need a formatted invoice, a pitch deck, or a budget spreadsheet? Grok generates production-ready Word docs, PowerPoint slides, Excel files, and PDFs, all styled and ready to share. You can also upload existing files and Grok will restructure, polish, or expand them.

Presentations

Build complete slide decks for pitches, quarterly reviews, and workshops with strong visual hierarchy and speaker notes.
“Design a Leadership Strategic Thinking Workshop deck with exercises, frameworks, and takeaways.”

Word Documents

Generate polished invoices, contracts, proposals, and resumes with full formatting — headings, tables, bullet points, and consistent styling.
“Create a professional enterprise invoice for APEX-2026-0547 with line items, tax, and payment terms.”

Spreadsheets

Generate formatted spreadsheets with formulas, color-coded sections, and clear structure — budgets, dashboards, and forecasts.
“Create a monthly budget tracker with income, expenses, and savings projections.”

PDF Reports

Create polished PDF reports, proposals, and research papers — formatted, styled, and ready to share.
“Generate a PDF report on AI interpretability formatted for conference submission.”

Build Your Own

Creating a skill takes seconds. Describe it in conversation, upload a file, or write one from scratch. Invested time perfecting a format or workflow? Just ask Grok to save everything it learned as a skill so you never have to explain it again.

Get Started

Skills are live now on Grok 4.3 across grok.com, iOS, and Android.

Original source

Introducing Grok Build

Early Beta

Now in early beta for SuperGrok Heavy subscribers — Grok Build is a new coding agent that runs right from your terminal.

Today we're launching an early beta of Grok Build, a powerful new coding agent and CLI for professional software engineering and complex coding work.

Available first for SuperGrok Heavy subscribers. Through this early beta, we will improve the model and product based on your feedback. Install Grok Build with a single command and sign in with your SuperGrok Heavy account:

curl -fsSL https://x.ai/cli/install.sh | bash

Plan, review, approve

For complex tasks, start Grok Build in plan mode. You can approve the plan, comment on individual steps, or rewrite it entirely before execution begins. Once a plan is approved, every change shows up as a clean diff.

Works with what you already use

Your AGENTS.md, plugins, hooks, skills, and MCP servers all work out of the box. Start Grok Build in your repo and it picks up your conventions instantly.

Subagents that work in parallel

For larger tasks, Grok Build delegates work to specialized subagents that run in parallel. Grok Build also supports deep worktree integrations, and you can launch subagents in their own worktrees.

Built to fit your workflow

Headless mode (-p) allows easily running agents inside scripts and automations. The CLI also provides full ACP support to build your own bots and agent orchestration apps.

Try it today

This is an early beta, and your feedback is the fastest way to make it better. Type /feedback in the CLI to send bugs, requests, and reactions straight to the team.

SuperGrok Heavy subscribers can install Grok Build with the command above and start building. If you're not on SuperGrok Heavy yet, upgrade here — we're excited to see what you create.

curl -fsSL https://x.ai/cli/install.sh | bash

Original source

Deep integrations that bring the apps you use every day directly into Grok.

Today, we're excited to launch Connectors — deep integrations that bring the apps you use every day directly into Grok. Connectors let Grok work with your tools end-to-end, turning fragmented workflows into seamless experiences. Read and summarize emails, update slides, organize your calendar, edit spreadsheets, and much more — all without constant copy-pasting.

Available on Grok Web today

Connectors are now live on Grok Web. Just sign in, click the + button, select Connectors, and click + Add connector to get started.

The following Connectors are now available:

• SharePoint: Grok can search, read, and synthesize the latest files, lists, and pages across your organization. With write permissions enabled, Grok can also create, edit, and update documents, lists, and pages. Advanced document editing capabilities are powered by Grok 4.3.
• Outlook: Search your inbox, calendar, and meetings with full context. Grok can draft and send emails, create calendar invites, or help you triage and prioritize your inbox.
• OneDrive: Instant access to all your personal and shared files. Ask Grok to analyze spreadsheets, presentations, reports, and more without ever leaving the chat.
• Google Workspace: Connect Gmail, Drive, Docs, Sheets, and Calendar with one integration. Grok can read and write emails, documents, spreadsheets, and events.
• Notion: Work directly in your Notion workspace. Search or edit pages, databases, and wikis across you and your team’s docs.
• GitHub: Connect your repositories, issues, and pull requests. Grok can search code, summarize PRs, review changes, and help you stay on top of your development workflow.
• Linear: Connect your Linear workspace to bring tasks, issues, roadmaps, and projects directly into Grok. Grok can search through your backlog, summarize sprint progress, draft updates, and create new issues.

In addition to all the connectors above, we’re also launching Bring Your Own MCP: Connect any custom Model Context Protocol server to Grok. No matter what your team uses - a homegrown knowledge base, proprietary APIs, or an internal MCP gateway - Grok can plug straight into your own tools and workflows.

To read documentation about connectors visit https://docs.x.ai/grok/connectors

What's coming next

We’ll be adding new Connectors regularly to support even more apps and workflows you use every day.

We’re also bringing full Connectors support to the Grok iOS and Android apps very soon, so you can access your connected tools wherever you are.

We can’t wait to see what you build with the connectors.

Original source

Higher realism. Stronger text rendering. Better creative control.

Quality Mode for image generation and editing is now live for enterprise developers and teams via the Grok Imagine API.

What Quality Mode Delivers

Photo of a woman with friends at the pool, wide shot, the woman is wearing a swimsuit that's all yellow toned and a matching cover up, looking relaxed, natural skin and pores, skin imperfections, Italian leisure aesthetic, warm afternoon sunlight, high-contrast dappled canopy shadows, deep forest green, terracotta orange, crisp canvas white, herringbone brick textures, medium format editorial photography, analogue film warmth, architectural framing, tranquil luxury vibe, deep depth of field.

Prompt: High end simple menu of french chocolate deserts

Prompt: Photo of [Influencer brewing morning coffee on tailgate with beautiful sunrise light], soft golden hour natural lighting with warm backlighting and subtle lens flares, warm earthy color palette with golden amber highlights and cool blue sky accents, serene contemplative atmosphere, realistic photographic rendering with natural material textures, balanced natural composition with leading lines, soft shadows and gentle highlights, crisp photorealistic details

Big leap in realism

Fine details, accurate textures, realistic characters and scenes.

Stronger text rendering

Clean, multilingual text capabilities.

Superior creative control

Tighter prompt following, deeper scene and world understanding, consistent brand results.

Leaderboard Rankings

xAI's Grok Imagine API ranks among the strongest models on independent leaderboards.

Text-to-Image Arena — Ranked by Lab (Top 5)

OpenAI
Google
xAI
Luma AI
Microsoft AI
xAI (Previous Model)
1398
1268
1223
1193
1181
1177

Source: LMArena Text-to-Image Arena as of
May 4, 2026.

Enterprise Use Cases

Quality Mode shines in business applications.

Product Visualization & Marketing Assets

Create photorealistic product renders that match your brand. Generate hero images, social assets, icons, and ad variations at scale.

PROMPT

Create an event ad for the Cybertruck in the first image. Maintain the same brand style as image 2 but make it Mars themed. Change the tagline to something different but maintain the tagline placement. Add the date and location of the event under the tagline: Mars, May 5th, 2035.

OUTPUT

Make Ideas Come to Life Faster

Visualize what you're thinking and share it with the team.

PROMPT

A small perfume bottle with a minimalist label that says "La Belle Vie", glass luxury perfume bottle with glass cap, standing on a stone villa window overlooking the ocean in warm sunlight, with Mediterranean retro-leisure aesthetic, green and brown hues against an expansive sea and wild flowers backdrop with hues of dark green, muted reds, vibrant greens, blues, pale yellows and natural retro film colors, light reflections cast from the bottle onto the stone, magazine style photo, quiet luxury aesthetic

PROMPT

Put this perfume bottle on a print pamphlet that a woman is holding in her hand, POV

OUTPUT

UGC Style Content

PROMPT

Full body composition, photorealistic portrait of a 25-year-old woman with long hair, natural realistic skin with visible pores and skin texture, streetwear vibe, wearing this exact black t-shirt with the precise white graphic logo from the reference image, fitted not oversized, black pants and all-black shoes, standing in confident pose with hand on hip, plain white seamless background, natural studio lighting, ultra realistic, high resolution, detailed

OUTPUT

PROMPT

Exact same woman, exact face, exact body, exact outfit from the reference image. UGC style, eating a piece of leftover birthday cake, picture from top down POV someone taking it, imperfect picture

OUTPUT

PROMPT

Exact same woman, exact outfit from the reference image. Mirror selfie in a reflective stainless steel elevator, holding a silver iPhone in one hand taking the photo. Soft natural lighting, casual streetwear aesthetic, realistic UGC photography style, relaxed lean against elevator wall with subtle smile

OUTPUT

Improved World Understanding

PROMPT:

Picture of a white frosted cake with a message on it in red frosting explaining in a nutshell who Alexander the Great is.

Broad Visual Range

From cinematic scenes to UI-style icons, Quality Mode can create any style.

PROMPT:

Cinematic shot of a couple on a picnic blanket in soft muted red and beige tones on hill, a picnic basket, wine, two glasses and book, soft diffused natural daylight, muted desaturated sage green and warm earth tone palette, gentle film grain, atmospheric haze, serene pastoral atmosphere, natural skin textures, intimate close-up or wide scenic composition, warm skin tones against cool backgrounds, contemplative mood

Quality Mode enhances Video Generation

Pairing our latest image model with our video capabilities creates endless possibilities for brands: social media video assets, product showcasing, ads and more.

How to Get Started

import xai_sdk
client = xai_sdk.Client()
response = client.image.sample(
    prompt = "A collage of London landmarks in a stenciled street-art style",
    model = "grok-imagine-image-quality",
)
print(response.url)

For full documentation, visit our docs page.

Welcome to the next level of Grok Imagine. Try Grok Imagine Quality Mode on API.

TRY QUALITY MODE

Original source

Your voice. Your brand. Clone a voice from a short recording and manage your entire voice catalog from the xAI console.

Today, we're introducing Custom Voices. Clone your voice from a few seconds of audio and use it instantly across Grok Text to Speech and Voice Agent APIs.

Alongside Custom Voices, the new Voice Library gives your team a single place to browse, preview, and manage all your voices from the xAI console.

Use Cases

Custom Voices unlock a new class of applications.

Brand Voice Agents

Give your customer support agent a consistent, recognizable voice that matches your brand identity, not a generic preset.

Custom Voices

Clone your voice in under two minutes. Use it everywhere.

Record about a minute of natural speech in the xAI console. Our pipeline verifies you're the voice owner, processes your recording, and delivers a production-ready voice model, all in under two minutes. Your custom voice inherits every TTS capability: speech tags, multilingual output, and both REST and WebSocket streaming.

Custom voices work everywhere our built-in voices do. Pass the voice_id to any TTS endpoint or use it with the Voice Agent API for real-time conversational agents.

Voice Safety

Every custom voice goes through a two-stage verification process before it can be created. First, the speaker reads a verification phrase that our STT engine transcribes and matches in real time, confirming intent and presence. Then we compute speaker embeddings from the verification clip and the full recording to confirm they belong to the same person.

You can't clone a voice from a pre-existing recording, and you can't clone someone else's voice.

Passphrase Check

Read a verification phrase aloud. Our STT engine transcribes and matches it in real time, verifying your consent and presence.

Speaker Similarity

Speaker embeddings from the passphrase and the full recording are compared to confirm they belong to the same person.

Voice Library

The Voice Library is a new section in the xAI console that organizes every voice available to your team, with your custom creations alongside our built-in voices. Browse, preview, and manage voices from a single page.

We've expanded our built-in voice catalog to over 80 voices across 28 languages. Listen to any voice across different scenarios before choosing one for your application.

There is no extra charge to use Text to Speech or Voice Agent APIs with custom voices.

Original source

Our most capable voice agent is now available via API.

Today, we're excited to announce a step change in xAI's Voice Agent capabilities: Introducing grok-voice-think-fast-1.0 — our new flagship voice model.

This new model excels at complex, ambiguous, multi-step workflows across customer support, sales, and enterprise applications. It is especially well-suited for high-stakes scenarios that demand precise data entry and high-volume tool calling to address the user's request.

Built for the messiness of the real world

We built grok-voice-think-fast-1.0 through tight collaboration with partners like Starlink to combine top-tier intelligence with low response latency and organic conversational ability.

Our model prioritizes snappy responses and unparalleled cost effectiveness without compromising on accuracy or tool orchestration. The result is a model that lets teams confidently deploy complex, multi-turn voice experiences across almost any conceivable use case: Customer support, phone sales, appointment booking, restaurant reservations, and more.

This new model takes the top spot on the τ-voice Bench leaderboard, which evaluates full-duplex voice agents under realistic conditions including noise, accents, interruptions, and turn-taking. See the benchmark details here.

The model has been battle-tested in the toughest real-world conditions: telephony audio, background noise, heavy accents, and frequent interruptions. It natively supports 25+ languages, making it ideal for global deployments.

Precise data entry and read-back

Collecting and confirming user information is critical for many workflows. Grok Voice is able to seamlessly collect email addresses, physical street addresses, phone numbers, full names, account numbers, and other structured data—even when information is spoken quickly or with a strong accent. It gracefully handles speech disfluencies and accepts natural corrections as a human would.

Real-time reasoning with zero added latency

Grok Voice Think Fast performs reasoning in the background, allowing it to think through challenging queries and workflows in real-time with no impact on response latency. This enables intelligent answers while retaining the dexterity needed for natural conversation.

Harder to fool

Voice models often default to confident, plausible-sounding answers, despite being completely wrong. We've built grok-voice-think-fast-1.0 to reason through edge cases before responding, catching obvious mistakes that other models get wrong.

Powering customer support and sales for Starlink

Grok Voice enables Starlink's phone sales and customer support experience at +1 (888) GO STARLINK. This requires working across numerous languages, helping customers through customer support scenarios, and onboarding new customers via sales:

• 20% conversion rate. In 1 out of every 5 of sales inquiries, the customer purchases Starlink service while on the phone with Grok.
• 70% resolution rate. The majority of customer support inquiries are resolved autonomously by the Grok Voice agent with no human in the loop.
• 28 tools. This single agent uses dozens of distinct tools across hundreds of support and sales workflows.
• Accuracy is critical. Grok handles high-stakes decisions; the model autonomously performs hardware troubleshooting workflows, issues hardware replacements, and grants service credits.

Original source

Today, we are excited to announce two powerful standalone audio APIs: Grok Speech to Text (STT) and Grok Text to Speech (TTS). Built on the same stack that powers Grok Voice, Tesla vehicles, and Starlink customer support.

These standalone endpoints make it straightforward for developers to integrate high-quality speech features into any application, whether you're creating voice agents, real-time transcription tools, accessibility solutions, podcasts, or interactive audio experiences.

Speech to Text

High accuracy, low latency.

• Generate transcripts from large audio files in milliseconds via our REST API
• Transcribe speech in real time with our lowest latency WebSocket API

We’ve added powerful features like word-level timestamps, speaker diarization, and multichannel support. It further includes intelligent Inverse Text Normalization that correctly handles numbers, dates, currencies, and more.

Pricing

We keep pricing straightforward and predictable: Speech to Text is $0.10 per hour for batch and $0.20 per hour for streaming. Full details and current rate limits are available in the xAI API console.

Enterprise-Grade Transcription

Grok STT is evaluated against the top commercial models on phone calls, meetings, video/podcasts, and telephony. It excels at entity recognition and business use cases like medical, legal, and financial.

Most transcription models give you raw spoken words. Grok Speech to Text goes further.

When you enable formatting, the API performs advanced Inverse Text Normalization that intelligently converts spoken language into proper structured output:

Multilingual fluency

The Grok Speech to Text API offers strong multilingual support across 25+ languages, switch languages seamlessly without missing a beat.

Multichannel & Diarization (Speaker Identification)

Transcribe multichannel audio files for perfect speaker separation with the same API.

Detect speakers in both pre-recorded and real-time streaming with word-level speaker IDs using Diarization.

Text to Speech

Fast, natural, and expressive voices with Speech Tags.

• Turn long-form text into speech with our REST API
• Generate speech in real time with our WebSocket API

Fine-Grained Control

Add natural prosody and emotion using simple inline and wrapping speech tags: [laugh], [sigh], [whisper], , , , and many more. These controls let you create engaging, lifelike delivery without complex markup.

Pricing

Text to Speech is priced at $4.20 per 1 million characters, with straightforward usage-based billing and no hidden fees.

Original source