Revenant XSpark, the team you may have seen competing in BGMI tournaments, is set to represent Southeast Asia at the Brawl Stars Challengers Finals 2026, where 12 of the world’s top regional teams will battle for the final qualification spot in the Brawl Stars Championship One Last Chance Qualifier (LCQ). The offline tournament will take place on September 5–6 at the ESA Espor Arena in Istanbul, Türkiye, with the winner earning a direct ticket to the LCQ in Guangzhou, China. Here’s everything you need to know.

Revenant XSpark lands in Group B alongside SK Gaming

The Challengers Finals feature 12 regional champions divided into four groups. Day one will consist of a round-robin group stage, with the top two teams from each group advancing to the playoffs. The tournament will then conclude with a single-elimination bracket and Grand Final on September 6.

Revenant XSpark has been drawn into Group B, where it will face SK Gaming from France and Good Game from South Korea in its opening matches. For the Indian organization, Istanbul represents the first major step toward another appearance at the Brawl Stars World Championship.

The team qualified through the Last Chance Qualifier before finishing 3rd–4th at the Brawl Stars World Finals 2025, becoming only the second Indian team to reach the World Stage Finals in the title. The roster is led by Ashmit “Sergeant Clash” Raj Singh, one of India’s most accomplished Brawl Stars players, alongside Southeast Asian veteran Jayden “X9Jay” Wong and Singaporean newcomer Walkthrough, who will make his international LAN debut in Istanbul.

Speaking on the matter, Rohit N. Jagasia, CEO of Revenant Xpark, said,

Our ambition is not simply to participate, it is to keep pushing deeper into the global competitive circuit. Istanbul is the first step, Guangzhou is the next target, and ultimately, we want to be back on the World Finals stage.

The Brawl Stars Challengers Finals will be streamed live on event.brawlstars.com, with the group stage scheduled for September 5 and the playoffs and Grand Final taking place on September 6.

Vodafone has announced Vodafone TV, a new entertainment hub that brings live television, streaming, gaming, music and apps together in one place.

The service is designed to simplify the increasingly messy experience of juggling multiple streaming subscriptions and devices. Vodafone TV will bring services including Netflix and HBO Max alongside Freely, more than 150 live channels, 300-plus cloud games and thousands of apps through the Google Play Store.

At the heart of the experience is a set-top box running Android TV. It supports 4K video, Dolby Vision and Dolby Atmos, while 3GB of RAM and 32GB of storage should give it enough power for navigating menus, streaming content and running games without feeling sluggish.

Vodafone is also putting more emphasis on discovery. Smart AI Search, Google Assistant and personalised recommendations are built into the platform, allowing users to search across different types of content without having to jump between individual services.

When enterprise buyers build out their next AI accelerator evaluation list this cycle, they’re more likely to put a non-Nvidia chip on it than Nvidia’s own next-generation GPU. According to VentureBeat’s July VB Pulse survey of 170 AI infrastructure respondents, 39.4% said they’re likely to evaluate non-Nvidia accelerators — AWS Trainium, Google TPU, AMD Instinct, Intel Gaudi or in-house ASICs — over the next 12 months, compared with 25.3% for Nvidia Blackwell (GB300) or other next-generation Nvidia GPUs, a 14-point gap.

Nvidia remains the default in most production environments. But organizations are building real optionality into their accelerator strategy rather than treating Nvidia as the only evaluation worth doing.

The finding sits inside a broader pattern: enterprises are expanding and optimizing the AI infrastructure they already operate before making another major platform change. Greater infrastructure activity did not produce greater urgency to switch platforms. The share of respondents expecting a platform change within three months fell from 38.3% in June to 28.8% in July, even as production adoption, accelerator utilization, and exploration of neoclouds and open-source infrastructure all rose.

Where is enterprise AI infrastructure actually growing?

The July data shows organizations operating AI infrastructure more intensively and putting more provider platforms into production.

Microsoft Azure posted the largest production adoption growth among the major platforms measured, with the share of respondents reporting Azure in production increasing from 29% in June to 47.1% in July, an 18.1 percentage-point increase. Some of that jump reflects who was surveyed: July’s respondent base skewed more up-market than June’s (57% at organizations above 1,000 employees, versus 37% in June), and Azure adoption rises with company size in both waves. Google’s Gemini was the most-used platform in both waves, with the share of respondents reporting it in production rising from 41.1% in June to 47.6% in July, narrowly ahead of Azure.

The share of respondents reporting OpenAI in production rose from 40.2% to 49.4%. Anthropic production adoption increased from 12.1% to 24.7%. 

Among enterprises that operate their own GPUs, the share running at half capacity or less fell from 83% in June (100 respondents) to 69% in July (155 respondents), with the share above 50% utilization rising from 13% to 23%.

The definition of infrastructure effectiveness is also becoming more operational. The share of respondents who selected uptime and reliability as important effectiveness measures increased from 42.1% to 51.2%. The share selecting throughput rose from 21.5% to 24.7%.

Ease of implementation improved from an average rating of 3.84 to 4.04 on a five-point scale. Overall satisfaction moved only slightly, from 4.07 to 4.14, while perceived value was essentially unchanged at approximately 3.9.

That combination is telling. Enterprises are not reporting a dramatic improvement in value simply because they are deploying more infrastructure. They are becoming more capable operators with better architectures, but they are also setting a higher bar for what that infrastructure must deliver, with reliability leading the way.

Why is platform-change urgency shifting outward?

The strongest counter-signal in the July findings is the declining share of respondents who plan to make an immediate platform change.

The share expecting a change within zero to three months declined by 9.5 percentage points. The share expecting a change within three to six months rose by 4.1 points, while the six-to-12-month window rose by 5.3 points. The share with no planned change remained effectively flat at approximately 40%.

Urgency is shifting outward, with the open-weight-model and open-source-harness debate playing a role in which pieces get enhanced versus fully replaced.

The selection criteria support that interpretation. Integration with existing cloud and data stack was the top factor in both waves, holding steady at 41.1% in June and 40.0% in July. The share of respondents prioritizing performance increased from 24.3% to 35.3%. The share prioritizing cost per million tokens increased from 7.5% to 15.9%, while the share prioritizing access to GPUs rose from 18.7% to 23.5%.

By contrast, the share selecting broad total cost of ownership as a leading factor fell from 34.6% to 21.8%.

The market appears to be moving from general infrastructure planning toward workload-level scrutiny. Buyers increasingly want to know how a platform performs under production inference, how reliably it operates and what each unit of useful work costs.

Interest in Nvidia alternatives is concentrated at the top

That 39.4% figure was 31.8% in June, already climbing before this wave. The alternatives enterprises are weighing include AWS Trainium, Google TPU, AMD Instinct, Intel Gaudi and other in-house ASICs.

Interest was even stronger among respondents with strategic purchasing authority, though the C-suite sample is small: the share of C-suite respondents likely to evaluate non-Nvidia accelerators rose from 42.9% (6 of 14) in June to 57.1% (12 of 21) in July. Among final decision-makers, the same interest rose from 35.4% to 50%.

This was especially true for organizations in the small and medium-size business tiers. Among organizations with 251 to 1,000 employees, the share increased from 41.4% to 53.2%. Among organizations with 101 to 250 employees, it rose from 33.3% to 57.7%.

These findings show organizations building optionality into their accelerator strategy.

The increased attention from C-suite respondents and final decision-makers suggests that accelerator diversity is becoming a strategic infrastructure question, not just a technical one for engineering teams.

Enterprises want to own the harness

The infrastructure findings align with a separate VB Pulse survey of agentic context layers. That survey included 101 substantive respondents in June and 101 respondents in July.

The AI harness is the operational layer connecting models to enterprise data, tools, orchestration, evaluation, identity, security, observability and business processes. It determines what an agent can access, which actions it can take and how the organization evaluates its output.

In July, 36.6% of context-layer respondents said they planned to retain best-of-breed standalone tools alongside their models. Another 36.6% expected to mix provider-native runtimes with standalone tools, while only 5.9% intended to build and own the context layer in-house.

Combined, 79.2% of July respondents favored an approach that maintained at least some architectural control outside a single model provider, compared with approximately 65.3% in June. Only 11.9% of July respondents favored consolidating onto a single model provider’s native context stack, down from 20.8% in June.

Most want to preserve provider choice, independent governance or control over critical components around the model.

The need for that control is becoming clearer. In July, 62.4% of context-layer respondents reported that a governed semantic or context layer was either in production or being built. Production adoption alone increased from 24.8% to 31.7%.

At the same time, 68.3% of July respondents reported experiencing at least one confident-but-wrong agent answer caused by missing or incorrect context, compared with 57.4% of June respondents.

The share expecting to use multiple retrieval architectures by use case increased from 12.9% to 28.7%. The share expecting to mix provider-native and standalone context tools increased from 20.8% to 36.6%.

The emerging architecture is a controlled combination of models, infrastructure, retrieval approaches, context systems and operational tooling selected by workload.

Are neoclouds gaining enterprise traction?

Neoclouds are specialized cloud providers focused heavily on AI infrastructure, particularly access to accelerators and supporting services. The July results suggest that these providers are becoming a more credible part of enterprise multi-provider strategies.

The share of respondents expecting to do more with neoclouds increased from 33% in June to 38% in July. At the same time, the share expecting to do less with neoclouds fell from 9.7% to 5.4%.

The movement was especially pronounced among respondents in the technology and software vertical. The share of that July segment expecting to do more with neoclouds reached 57.6%, compared with 44.4% in June.

Current production adoption remains much smaller than broad expansion intent. Across the named providers measured consistently in both waves, such as CoreWeave, Lambda, Crusoe and Nebius, production use increased from 1.9% of June respondents to 5.9% of July respondents.

The difference between 38% expansion intent and 5.9% current named-provider production use may point to a sizable evaluation and adoption pipeline.

The neocloud demand pipeline is not theoretical. CoreWeave reported around $104 billion in revenue backlog at the end of June, excluding more than $25 billion in additional customer commitments secured during early Q3. Nebius does not disclose a directly comparable backlog metric, but said it could sell its entire 2027 capacity under current terms and reported four second-quarter AI cloud agreements, each averaging more than $1 billion in total contract value.

The larger implication is that neoclouds are becoming a viable source of strategic leverage. They give organizations additional options for accelerator availability, software stacks, workload placement and ammunition for negotiations with hyperscale providers.

Neoclouds will still have to demonstrate enterprise-grade reliability, security, support, networking, and data management capabilities. Specialized compute access may open the door, but durable enterprise adoption will depend on the surrounding operational stack. 

Is open-source AI infrastructure usage growing?

The most accurate answer is that open-source production usage is growing, while broad platform consideration remains relatively flat.

The share of respondents reporting a custom, self-managed open-source production stack increased from 3.7% in June to 12.9% in July. The stack definition included technologies such as PyTorch, Triton, vLLM, Ray and Kubernetes.

The movement was visible across several segments with July bases above 20 respondents:

• Among individual contributors, 23.9% reported production use in July.

• Among recommenders and influencers, 13.7% reported production use in July.

• Among organizations with 251 to 1,000 employees, 12.8% reported production use in July.

The share of respondents using open-source key-value cache tooling, including LMCache and vLLM prefix caching, increased from 6.5% to 11.8%. Among technology and software respondents, usage increased from effectively 0% to 13.3%.

Open-source platform consideration ticked up slightly but remained essentially unchanged, moving from 5.6% to 6.5%.

This combination suggests that growth is concentrated among organizations moving into implementation rather than across a dramatically larger population of evaluators. Open source appears to be deepening inside an active portion of the market.

Organizations may be turning to open-source components for greater portability, model choice and control over inference optimization. But ownership also transfers responsibility. Teams adopting self-managed stacks must operate upgrades, security, observability, integration and production support themselves.

That combination of more activity, less urgency and more optionality is the throughline across all of it. Enterprises are running more AI infrastructure while deliberately keeping multiple paths open on chips, clouds and the layer that connects models to their own data. The next platform change, when it comes, will be a choice made from a stronger position.

Notes on methodology

For this article, I compared two independent, cross-sectional infrastructure survey waves: 107 respondents in June 2026 and 170 respondents in July 2026. These waves are not a longitudinal panel, so the findings describe changes between respondent populations rather than changes made by the same organizations. Platform-change timing shares add to slightly more than 100% because a small number of respondents selected more than one window (5 in June, 9 in July).

Sample composition changed between the waves. Respondents selecting the 1–100 employee organization-size category were excluded before calculating results. The remaining wave composition still differed, including a larger July share from organizations with more than 10,000 employees. Month-to-month movements should therefore be treated as directional signals rather than proof of causation. No statistical-significance testing was applied to the comparisons reported here.

The context-layer findings come from a separate survey, with 101 substantive respondents in June and 101 in July. Those results use a different respondent base and are included as supporting evidence, not combined with the infrastructure-survey results.

This is obviously intended for the competitive esports market.

The timing of newly available scans—typically within a day, if not hours, of me and a small sample of other victims presenting them at rental companies or others—likely means that Nexus has near real-time access to data flowing through the third-party scanning service these businesses are using. Over a span of 24 hours, Krebs said the number of driver’s licenses listed as available grew by almost 400,000. That’s another indication that the breach has been ongoing and new cards become available shortly after they’re harvested.

Using publicly available information, Krebs found that IDScan.net, a New Orleans-based ID scanning service, has announced an exclusive arrangement with Planet13. It also listed Hertz and 11 other companies as using its services. IDScan.net went on to say that its scans capture both infrared and ultraviolet spectra.

Representatives from IDScan didn’t immediately answer questions sent by email. An IDScan.net spokesperson told Krebs the company is investigating. My car rental company representatives also didn’t immediately answer questions.

The availability of my driver’s license to anyone willing to cough up a fee isn’t exactly a comforting thought. Yes, my personal details—including current and former addresses, Social Security number, demographics, and more—have been breached before, just as they have for millions, if not billions, of others around the world.

This dump is more troubling because of the purported availability of scans in ultraviolet and infrared. Fortunately, Nexus went dark within hours of the KrebsOnSecurity scoop, although that also means there’s no way for people to check if their IDs are included. Also somewhat consoling is the ongoing investigation by the FBI.

INDIANAPOLIS — There’s the mother of two who needs a full-time job to keep up with the bills, but can’t afford full-time childcare. There’s the medical technician who stopped buying groceries and turned to food banks so she could pay for her infant’s care.

There’s the baby who remained in foster care, even after his mother completed the steps to bring him home. She had found a job, as the state required, but couldn’t afford his childcare.

It’s hardly news that childcare is expensive, straining the budget of even middle-class parents. But these families all qualify for government-funded childcare assistance, intended to drastically reduce the childcare bills of needy parents so they can work or go to school. Instead, Indiana’s program, which had run short on funds, waitlisted them.

Hundreds of thousands of children were sitting on childcare assistance waitlists in 23 states and the District of Columbia as of this spring, according to an analysis by The Associated Press. In three more states, many eligible families who applied were simply turned away.

Childcare assistance, funded by the federal government with help from states, is supposed to aid working parents who are low-income, homeless or caring for foster children. But waitlists have ballooned since 2023, when eight states had them, according to the AP analysis. Reporters collected waitlist and enrollment data from 47 states and the District of Columbia.

Many of the waitlists first started after $28 billion in pandemic aid expired in the fall of 2024. The Republican-led Congress declined to extend the extra money. Some states have increased funding to try to fill the gap, but families’ need for childcare assistance has only grown, as they’ve faced mounting costs for food and gas.

While eligible families sit on waitlists, untold numbers of adults are sidelined from education or the workforce because they can’t afford childcare. Those parents who remain in school or keep working make difficult tradeoffs. In interviews, parents described forgoing necessities, falling behind on bills or relying on acquaintances to watch their children. Unable to find viable childcare, some have resorted to bringing their children with them to work or class.

For Meygan Maloney’s family in rural Indiana, waiting for childcare assistance means month after month of falling short financially. Maloney was working as a caregiver for disabled adults before she had her second child in 2023. Around the same time, the family took emergency custody of a close relative’s newborn daughter.

The next three years were marked by deepening financial struggles. Maloney only received $300 a month to care for her foster daughter, and both toddlers were put on a waitlist when she applied for childcare vouchers for 2025.

While the girl returned to her family in the fall, Maloney and her husband are still behind on bills. They recently refinanced their home, and her husband is working seven days a week as a tire technician and delivery driver. A local childcare provider gave her a steep discount, allowing her to leave her son four mornings a week to clean at a local hospital.

“I felt useless as a stay-at-home mom. I felt like I wasn’t contributing to our household,” Maloney said, her voice cracking. “We’re trying to get out. But for us to get out, we need the assistance.”

Foster parents left to pay for childcare in Indiana

Indiana started its waitlist in December 2024, a few months after pandemic aid ran out and it became clear the state would struggle to support families already in the program. Until May of this year, it put virtually everyone who applied on the waitlist, including foster parents. By spring 2026, nearly 37,000 children were waiting for assistance.

Maloney and other foster families feel like they were misled by the state, which had for many years paid for childcare for foster children. Malinda Cox, an Indianapolis mom who took in a newborn last year, said she was forced to dip into her family’s savings to pay for the baby boy’s childcare so she could return to work. At one point, she considered giving up custody of the boy because her family could not afford his childcare bill.

In the meantime, the boy’s biological mother was completing the requirements she had to satisfy to take him home. She had taken parenting classes and gotten a job. But she could not afford daycare, and because the boy could not get childcare assistance, he had to remain with Cox.

“We started to feel, like, guilty. He should be home with his mom, but (because of) this massive system upset, he’s not,” Cox said.

In April, Indiana Gov. Mike Braun, a Republican, announced the state would send $200 million in surplus funds to the state’s childcare assistance program, which would allow it to move roughly 8,000 children off the waitlist.

Even before the recent growth in waitlists, federal funding for childcare assistance has always fallen short of providing for every needy family, only providing enough money to serve a fraction of those eligible. Experts say waitlists represent the tip of the iceberg, because many eligible families who need the help don’t know about the program or haven’t applied, daunted by tales of long waitlists.

The origins of the funding shortfall are cultural, said Ruth Friedman, who headed the Office of Child Care under Democratic President Joe Biden. For decades, “childcare was the responsibility of the family,” she said. “The system is fundamentally broken, and if the government doesn’t help fix the system … it will not be affordable for families.”

There are stark differences of opinion on whether the government should provide help — or how much. This year, 40 congressional Republicans called for “robust” funding of the federal program that underwrites childcare assistance. Instead, it received the same money as the year before. President Donald Trump said during his campaign that he could make childcare more affordable by using tariff revenue, but this spring he told a White House audience that paying for childcare should be up to states.

“We’re fighting wars. We can’t take care of daycare. You got to let a state take care of daycare, and they should pay for it, too,” Trump told people gathered for an Easter luncheon.

His administration has largely focused on allegations of fraud in the program, at one point halting funding to five Democratic-led states. After a lawsuit, a judge ordered Trump’s administration to restart the funding.

Moms bring babies to college or work

Without affordable childcare options, parents have resorted to bringing their children with them to work or school.

In Minnesota, Deaira Gresham’s 1-year-old daughter Seaira has been on the waitlist for childcare assistance since birth, one of 9,000 children in her state awaiting aid as of this spring. Gresham, who works as an in-home health aide and takes a full load of classes at a chiropractor school, had some help with Seaira from her mother. But oftentimes, she was forced to take Seaira with her to class and to the homes of her clients. It took a toll — Gresham failed several classes juggling school and caring for a baby.

This fall, she lucked out, getting a childcare grant for parenting students, allowing her to put Seaira in full-time care.

“Parents are in desperate need,” Gresham said. “We need that break. We need to be able to go to school, go to work and know that our kids are in good hands.”

It’s not just their work or college classes that parents worry about. Research has shown that high-quality childcare can boost a child’s social-emotional and academic development.

Carmen Perez, who lives north of San Francisco, has applied for childcare help for all four of her children. Her eldest son never received it before aging out of the program. He struggled academically in ways that his sisters, who attended subsidized preschool, did not.

Now her two-year-old son is on a waitlist. She fears he will face the same challenges as his older brother. “It makes me sick.”

On a rainy May day in Indianapolis, Cox bundled up her foster son and delivered him to a social worker. She packed up her own car with the boy’s things — clothes, diapers and a small changing table. The boy’s biological mother still had not gotten childcare assistance, but Cox had negotiated a deal with a childcare provider to offer a massive discount. When even that was out of the mother’s reach, Cox agreed to pay half.

She arrived at the boy’s new home to watch him reunite with his mother. She felt a mix of grief over parting with him and elation over seeing his delight with his mother.

Three weeks later, the mother got notice: He had finally been approved for a voucher.

Anthropic just rolled out a genuinely useful update. Computer use in Claude Cowork and Claude Code now lets Claude work in the background on your Mac, meaning it clicks, types, and opens apps while you do literally anything else.

How does Claude work in the background without taking over my screen?

Claude now runs in a background window instead of taking over your display. That means you can keep browsing, writing, or coding while Claude quietly handles its own task behind the scenes.

It won’t grab your mouse or keyboard mid-task, and it waits if you’re in the middle of typing. The only time it interrupts you is when a task needs the full screen, and even then, it asks permission first, just once per session.

What tool does Claude actually use to get things done?

Claude doesn’t jump straight to clicking around your screen. It first checks if a connector like Gmail, Google Drive, or Slack can handle the job, since that’s the fastest route. If there’s no connector, it tries your browser next, either the one built into Claude Desktop or your own Chrome browser through Claude in Chrome. 

Only when neither option works does Claude fall back to navigating your screen directly, handy for internal dashboards or specialized tools at work without a proper connector.

What can you actually hand off?

There’s plenty of busy work you can hand off to Claude. You could ask Claude to pull together a competitive analysis from local files and connected tools, then format it into a report. Or have it open your phone simulator, poke around the app you’re building, and flag any UX issues. And since Claude works in the background, if your task depends on a physical machine, it will keep chugging along even after you walk away from your desk, as long as your computer stays on.

The feature is currently in beta for Pro and Max plans and is available in Cowork and Claude Code on both macOS and Windows. Head to Settings → General → Computer use to turn it on, and make sure you read up on using Cowork safely before you let Claude loose on your desktop.

Meta is entering the increasingly competitive real-time speech-to-text market with Muse Voice Transcribe, a new audio perception model that combines streaming transcription, endpoint detection and speaker diarization for more than 20 speakers — at a public API price of just $0.18 per hour of processed audio.

Developed by Meta Superintelligence Labs, Muse is designed to process speech while it happens rather than waiting for a recording to finish. Meta’s launch post for Muse Voice Transcribe says the model supports long audio exceeding an hour, seamless multilingual code-switching, language and keyword biasing, and diarization without a separate post-processing pipeline. The model was trained across more than 70 languages, with 25 extensively validated for the initial release.

The 20-plus-speaker figure is substantial, but it is not a world record. A review of current vendor documentation turns up systems with higher published ceilings. Speechmatics’ real-time transcription service says it can identify 50 speakers by default and up to 100 when the limit is increased, while Amazon Transcribe’s diarization documentation specifies a maximum of 30 unique speakers, including for streaming transcription. (Speechmatics)

Muse nevertheless lands toward the high end of the market, and Meta’s broader proposition is arguably more important than the raw maximum: high-capacity real-time diarization combined with low-latency transcription, endpointing, multilingual code-switching and aggressive API pricing in the same model.

For enterprise developers building meeting systems, call analytics, live assistants or ambient AI, that combination could matter more than who holds the speaker-count record.

Diarization is becoming part of the core voice stack

Traditional speech recognition answers a relatively simple question: What was said? Diarization adds another: Who said it?

That distinction becomes critical as transcripts feed downstream AI systems. A meeting assistant can correctly transcribe every sentence and still create an unreliable corporate record if it attributes an approval, commitment or objection to the wrong participant. The same issue affects customer-service analytics, compliance workflows and AI agents operating in rooms where several people can speak.

Muse incorporates speaker attribution directly into its autoregressive multimodal architecture. Meta says audio arrives in 80-millisecond chunks, or 12.5 chunks per second, with each transformed into a soft token. At each step, the model decides whether to consume more audio or emit text. Meta calls this mechanism adaptive delay: rather than applying one latency budget to every word, Muse can wait longer when speech is ambiguous and commit earlier when it has enough context. Meta says reinforcement learning combines word-error-rate and delay rewards to train that behavior. Meta’s technical explanation of Muse details the architecture. (Meta AI Research)

Speaker attribution and endpointing then become part of the same token sequence. A <|start_of_turn|> token marks a potential new speaker turn, tokens such as <|speaker_A|> identify the speaker, and separate onset and endpoint tokens identify speech boundaries. Meta says it trains ASR, diarization and endpointing together rather than running speaker clustering as an unrelated downstream process.

Meta’s Model API speech-to-text documentation also exposes diarization as a first-class operating mode alongside push-to-talk and endpointing. Speaker labels such as A and B are scoped to a session rather than verified identities, and the API provides turn-level rather than word-level timestamps.

20+ speakers is high, but Speechmatics goes considerably higher

Speaker-count comparisons require care because vendors implement diarization differently and do not all publish a maximum.

Speechmatics currently makes the strongest explicit real-time capacity claim found in this review. Its real-time STT documentation says speaker diarization is available live, while its real-time FAQ says the system supports 50 speakers by default and can be increased to 100.

AWS likewise exceeds Meta’s stated figure: Amazon Transcribe can differentiate a maximum of 30 unique speakers, and AWS provides explicit instructions for speaker partitioning in a streaming transcription.

Soniox supports diarization in both real-time and asynchronous processing, but documents a maximum of 15 speakers per session. AssemblyAI’s streaming diarization system lets developers set max_speakers between one and 10. Both companies caution that live speaker attribution is more difficult because streaming systems must make decisions with less future audio context than offline models.

xAI’s current Speech-to-Text API also supports speaker diarization in streaming mode, but its documentation reviewed for this story does not publish a maximum diarized-speaker count, so a direct ceiling comparison with Muse is not possible. (X.ai Docs)

That means it would be inaccurate to describe Muse’s 20-plus capability as a new global record. The highest explicitly documented real-time number identified in this survey is Speechmatics’ configurable 100-speaker ceiling.

Meta also does not demonstrate 20-plus simultaneous participants in its launch material. Its principal live demonstration uses eight speakers, while its long-form recording contains 11 labeled participants. The 20-plus number is a stated model capability rather than the participant count in the public demos.

At $0.18 per hour, Muse competes aggressively on price

Meta’s pricing makes the competitive picture more interesting.

According to its Muse Voice Transcribe developer page, Muse costs $3 per 1,000 minutes, or $0.18 per hour. Streaming and non-streaming transcription cost the same, and Meta says zero-data-retention processing is priced at parity with standard processing. Billing applies to audio actually processed and is rounded down to whole seconds.

Standardizing publicly posted rates to one hour of streaming audio gives the following rough comparison:

The comparison is necessarily imperfect. Qwen’s price varies by deployment geography; its international real-time rate of $0.00009 per second works out to about $0.324 per hour. Google’s Gemini figure is an estimated blended token cost rather than a flat hourly tariff. AWS prices vary by region and usage tier. ElevenLabs lists $0.39 per hour on its API pricing page but advertises $0.28 per hour or lower on annual Business plans.

Deepgram’s pricing particularly illustrates why feature-level comparisons matter: its current Nova-3 Multilingual streaming rate is about $0.35 per hour, but speaker diarization costs another $0.002 per minute, bringing the comparable total to roughly $0.47 per hour. AssemblyAI similarly lists $0.45 per hour for Universal-3.5 Pro Realtime and another $0.12 per hour for streaming diarization.

Cartesia is harder to normalize because Ink-2 is packaged through monthly credit plans rather than a simple metered PAYG hourly rate. Its $5 Pro plan includes roughly nine hours and 16 minutes of Ink-2 transcription, which works out to about $0.54 per transcription hour if every credit is consumed exclusively on STT. That should not be treated as equivalent to a standalone $0.54 hourly API tariff.

Even with those caveats, Muse’s positioning is clear. It is not the absolute cheapest streaming transcription service — Soniox currently publishes a lower equivalent rate — but $0.18 per hour with diarization included puts Meta toward the low end of the market, especially against providers that charge separately for speaker attribution.

At 1,000 hours of processed audio, Meta’s public rate implies roughly $180 in transcription charges.

Meta also leads its launch accuracy benchmarks

Price matters less if it comes with a large accuracy penalty. Meta’s benchmark material argues the opposite.

On the Artificial Analysis AA-WER Streaming Index supplied with the launch, Muse records a 3.1% final-transcription word error rate, ahead of Cartesia Ink-2 at 3.4%, ElevenLabs Scribe v2 Realtime at 3.6%, Qwen3 ASR Flash Realtime at 3.7%, GPT Live Transcribe and Grok Speech to Text Streaming at 3.9%, and Gemini 3.5 Transcribe Live and AssemblyAI U3.5 Realtime Pro at 4.0%.

Meta points out that Muse took the number one spot on third-party independent AI benchmarking firm Artificial Analysis’ streaming speech-to-text evaluation as of September 1. Meta published the following benchmark charts in its launch post.

Its diarization result may be even more relevant to the product’s positioning. Meta reports an average 17.5% diarization error rate across AMI-IHM, AMI-SDM and VoxConverse, lower than the competing systems shown in its chart.

Speaker capacity and diarization error rate should not be conflated. A platform capable of representing 100 people is not automatically better at correctly attributing speech than one supporting 20, and Meta’s benchmark does not test every competitor operating at its advertised maximum speaker count.

There are deployment tradeoffs as well. Meta’s API currently provides turn-level but not word-level timestamps, and it does not expose word-level confidence scores, sound-event detection or emotion detection. The documentation also specifies eight concurrent streams per tenant by default and real-time sessions of up to 60 minutes before an application must reconnect.

Still, Muse’s launch creates an unusually sharp price-performance proposition. Its 20-plus-speaker diarization does not establish a world record, but the record may be the less important metric. For enterprise developers, the larger question is whether a service can preserve speaker attribution, accurate text and usable turn boundaries while a complicated real-world conversation is still unfolding.

At $0.18 per hour, with 20-plus-speaker diarization inside the same real-time model that currently leads Meta’s supplied streaming accuracy benchmarks, Muse Voice Transcribe gives enterprise teams a serious new option for meeting intelligence, live transcription and voice-agent infrastructure — while putting additional pressure on competitors to compete on speaker-aware accuracy and total operating cost, not merely raw speech recognition.