Sign in

Inworld Realtime TTS-2

by Inworld AI Model API in Text-to-speech

Hosted

Theai, Inc. dba Inworld AI · inworld.ai · status page · who's behind it

Inworld's hosted text-to-speech models, inworld-tts-2 and inworld-tts-2-flash, over REST, HTTP streaming, WebSocket, async jobs and an OpenAI-compatible endpoint. TTS-2 takes bracketed steering, and an opt-in Reasoning step writes the delivery direction itself.

Best for Agents that generate replies or stories at runtime and want delivery chosen from context without writing tags, and for per-message or pre-generated audio.

Write a review Compare Follow changes

Viewed under 500 times in the last 30 days · See the full traffic report

Claim this listing

Is this your product? Claim it to see how agents find it: lookups through the API, MCP and letme, the comparisons it's in, and what people search for in TTS. Free, and it never changes the grade. Or verify it, or log in if it's already claimed.

Assessment. TTS-2 takes bracketed steering, and opt-in Reasoning writes a delivery direction from the text and up to 600 characters of context, returning it in the response. Per-character prices are public and the free plan needs no card. No OpenAPI file was found, keys can't be limited to TTS, and the TTS-2 page's #1 arena claim didn't hold on 30 October.

Facts

Access how an agent reaches and pays for it
Transport
HTTP, websocket
Endpoint
https://api.inworld.ai
Auth
API key
Pricing
Freemium · $25 / mo
x402
No
Licence
Proprietary service under Inworld's Terms of Service. The Node.js SDK `@inworld/tts` and the CLI `@inworld/cli` are MIT on npm
Where it lives packages, source, docs and releases
Packages
npm @inworld/tts
pypi inworld-tts
npm @inworld/cli
llms.txt
published
Last release
npm / week
3.6k
From our research 15 facts checked against the vendor's pages
Surface listed
Inworld's TTS API at https://api.inworld.ai with the inworld-tts-2 and inworld-tts-2-flash models, read from docs.inworld.ai, the launch posts and the pricing, legal and security pages on 30 October 2026. No account was created and no API call was made
Models
inworld-tts-2 (steering, Reasoning, professional voice clones, 100 ms P90 server-side time to first byte per Inworld) and inworld-tts-2-flash (no steering, 20 ms per Inworld). inworld-tts-1.5-max and -mini are deprecated, and inworld-tts-1 and -1-max were discontinued on 15 June 2026 with requests routed to newer models
Endpoints
POST /tts/v1/voice (up to 2,000 UTF-16 units, base64 audio in JSON, 16 MB output cap), POST /tts/v1/voice:stream (up to 4,000, NDJSON), WebSocket /v1/tts/synthesize:websocket, POST /tts/v1/voice:synthesizeAsync (up to 100,000 characters, Preview), voice:synthesizeBatch (Preview) and OpenAI-compatible POST /v1/audio/speech
Reasoning
synthesisContext.reasoning (REASONING_ON), optional reasoningContext up to 600 characters, never spoken and not billed, and previousRequests for earlier lines. The response carries reasoningInstruction, on the first message when streaming. An inline tag overrides it from where it appears. If no instruction is ready in time the text is synthesised without one, and synchronous requests start slightly later with it on. Launched 28 October 2026
Controls
Inline bracketed steering and non-verbals, [reset], deliveryMode (STABLE, BALANCED, CREATIVE) on TTS-2, SSML break tags, inline IPA, named pronunciation dictionaries, language tags, verbatim tags, text normalisation on or off, and word or character timestamps with visemes
Output
MP3 (default), LINEAR16, PCM, WAV, OGG_OPUS, FLAC, MULAW and ALAW, 8 to 48 kHz
Languages
623 language and locale codes covering 470 languages in three tiers. Tier 1 is 15 languages, Tier 2 is 79 and Tier 3 is 376 best-effort languages that Inworld hasn't evaluated
Limits
Concurrent generations 5 on On-Demand, 10 Creator, 50 Builder, 150 Developer, 500 Growth. WebSocket connections 50 to 5,000. Voice cloning and design 2 to 20 requests a minute. Limits are per account and shared across keys
Errors
gRPC-style JSON with code, message and details. The reference shows one 4XX example (code 5, unknown voice), and the concurrency page documents code 8 ResourceExhausted. 429 is documented with exponential backoff and jitter guidance. No full TTS error list was found
Credentials
Workspace API keys in Authorization: Basic, deleted to revoke, with optional write toggles for the Voice and Router APIs and a separate Realtime-only key type. One-time tokens (single use, 15 minutes by default) for clients. Session tokens are deprecated
Plans
On-Demand free, Creator $25, Builder $100, Developer $300 and Growth $1,500 a month, each with credits equal to the price that roll over for 3 months, and Enterprise custom with an SLA, DPA and EU or India residency
Data handling
Zero data retention covers TTS text and audio on Enterprise, or as a Growth add-on. Usage metadata, voice cloning samples and voice design prompts are kept even then. Async and batch results stay in Inworld storage for about 7 days unless delivered to the customer's bucket
Certifications
SOC 2 Type II, HIPAA and GDPR claimed on the security page, with details on a trust centre at trust.inworld.ai that wasn't read
Other Inworld products
The same key and docs cover STT, the Realtime speech-to-speech API, voice cloning and design, and Inworld Inference for LLMs. The ultravox listing records Inworld's announcement of 30 September 2026 that it acquired Ultravox
Independent ranking
On the Artificial Analysis provider-voice leaderboard, read 30 October 2026, Realtime TTS-2 was 8th (Elo 1,267, 2,527 appearances) and Realtime TTS-2 Flash 16th (Elo 1,217). Eleven v4 Turbo led at 1,336

Facts verified 30 October 2026 from vendor docs, repositories and package registries. JSON · Markdown

Strengths

  • Reasoning returns the direction it applied as reasoningInstruction, so an agent can log it and reuse it as an inline tag
  • Per-character prices are public, from $25 per million characters for TTS-2 on the free plan to $12.50 on Growth
  • REST, HTTP streaming, WebSocket, async jobs and an OpenAI-compatible /v1/audio/speech endpoint
  • llms.txt, Markdown twins of every page and a keyless docs MCP server at docs.inworld.ai/mcp
  • Single-use client tokens, 15 minutes by default, keep the API key off browsers and phones

Weaknesses

  • No OpenAPI or other machine-readable contract was found for the TTS API
  • API keys can't be limited to TTS. Write toggles cover only the Voice and Router APIs
  • The 28 October model update replaced inworld-tts-2 in place, with no dated snapshot to pin
  • Zero data retention is Enterprise only or a Growth add-on, and the SLA is Enterprise only
  • The TTS-2 page claims #1 on the Artificial Analysis Speech Arena, which ranked it 8th on 30 October 2026

Before you call it notes for agents

  1. Send the copied Base64 credential as-is in Authorization: Basic. Don't encode it again.
  2. Set modelId explicitly, because SDK defaults may name deprecated models.
  3. Keep text under 2,000 UTF-16 units on /tts/v1/voice and 4,000 on voice:stream, or use an async job.
  4. Turn on Reasoning with synthesisContext.reasoning: REASONING_ON and store reasoningInstruction to repeat a delivery as a bracketed tag.
  5. On 429 or error code 8, back off exponentially with jitter. No Retry-After header is documented.

Who's behind it provenance 75/100

  • Legal entity namedTheai, Inc. dba Inworld AI20/20
  • Domain ageinworld.ai, no registry record we could read0/15
  • Endpoint on the vendor's domainapi.inworld.ai15/15
  • Terms of serviceread, states 6 of the 7 things a reader expects9.1/10
  • Privacy policyread, states 7 of the 8 things a reader expects9.3/10
  • Status pagestatus.inworld.ai10/10
  • Changelogpublished10/10
  • security.txtnot found0/10
  • People named3 named with their role, none with a public profile7/10
  • Support routea support address, help pages, a community5/5

Terms and privacy, as read

Terms of service dated 2025-06-11, states 6 of 7

TL;DR Dated 2025-06-11. States 6 of the 7 things a reader expects, and we didn't find a service level. The rules found no clause to flag.

Gives the date it was last updated Last updated 2025-06-11
Effective Date: June 11, 2025

Without a date nobody can tell which version they agreed to.

Names the governing law or courts Disputes go to the courts of Santa Clara, California
Except as provided in the dispute resolution section above, all claims arising out of or relating to these Terms will be brought exclusively in the federal or state courts of Santa Clara, California.

Says where a dispute would be heard and under whose law.

States a limit on its liability Capped at the greater of $100 and the fees paid in the 12 months before the claim
OUR AGGREGATE LIABILITY UNDER THESE TERMS WILL NOT EXCEED THE GREATER OF: (A) THE AMOUNT YOU PAID FOR THE SERVICE THAT GAVE RISE TO THE CLAIM DURING THE 12 MONTHS BEFORE THE LIABILITY AROSE OR (B) ONE HUNDRED DOLLARS ($100).

Says the most the vendor would owe if the service causes a loss.

Says how the agreement or account can be ended
If your payment cannot be completed, we may downgrade your account or suspend your access to our Services until payment is received.

Says when the vendor can cut off access and what notice it gives.

Says how changes to the terms are announced Gives 30 days of notice before a change
If we increase our subscription prices, we will give you at least 30 days' notice and any price increase will take effect on your next renewal so that you can cancel if you do not agree to the price increase.

Says whether a customer hears about a change before it binds them.

Lists what users may not do
YOU ACCEPT AND AGREE THAT ANY USE OF OUTPUTS FROM OUR SERVICE IS AT YOUR SOLE RISK AND YOU WILL NOT RELY ON OUTPUT AS A SOURCE OF TRUTH OR FACTUAL INFORMATION, OR AS A SUBSTITUTE FOR PROFESSIONAL ADVICE.

The acceptable-use rules an agent acting for a user has to stay inside.

Refers to a service level or uptime commitment

Not found in the text.

Says whether availability is promised and where the promise is written.

The document · read 2026-10-30 · 2,623 words

Privacy policy dated 2026-01-18, states 7 of 8, 1 to know

TL;DR Dated 2026-01-18. States 7 of the 8 things a reader expects, and we didn't find where data goes. To know before relying on it, selling or sharing data for advertising.

Says it sells personal data or shares it for advertising
Some of the activities described in this section may constitute “targeted advertising,” “sharing,” or “selling” under certain laws.

Personal data is passed to advertising partners, or the document says its sharing may count as a sale under privacy law.

Gives the date it was last updated Last updated 2026-01-18
Last Updated: January 18, 2026

Without a date nobody can tell which version applied when data was collected.

Says what personal data is collected
This Privacy Notice applies to information we collect when you access or use our websites and any related services (collectively, the “Services”), or when you otherwise interact with us, such as through our customer support channels.

The basic statement a privacy policy exists to make.

Says how long data is kept For as long as needed, with no period named
We store personal information for as long as necessary to carry out the purposes for which we originally collected it, satisfy our legal requirements, or for other business purposes explained in this Privacy Notice.

Says when data sent to the service is deleted.

Says who else receives the data
Vendors: we disclose information to vendors, service providers, contractors and consultants that need this information to provide services to us, such as companies that assist us with web hosting, payment processing, fraud prevention, customer service, data enrichment, analytics, and marketing and advertising.

Names the sub-processors or service providers the data is passed to, or where they are listed.

Says whether personal data is sold or shared for advertising Says it does not sell personal data
You can also adjust certain cookie settings by selecting “Do Not Sell or Share My Personal Information” within the cookie icon at the bottom-left corner of our website.

A plain statement either way.

Says what rights people have over their data
You have the right to opt out of sharing, sales, and targeted advertising by:

Access, correction, deletion and objection, and how to use them.

Gives a privacy contact privacy@inworld.ai
Depending on where you reside, you may opt out of these disclosures by emailing privacy@inworld.ai.

An address or officer to send a request to.

Says where data is transferred or stored

Not found in the text.

The countries data goes to and the safeguard used.

The document · read 2026-10-30 · 3,236 words

A reading by a fixed set of rules, each answered with the vendor's own sentence. It isn't legal advice, a rule can miss a clause or misread one, and the document itself is what binds. How it's read and scored.

The Terms of Service (effective 11 June 2025) are an agreement with Theai, Inc. dba Inworld AI under California law, with courts in Santa Clara. The Privacy Notice (last updated 18 January 2026) names Theai, Inc.

inworld.ai/.well-known/security.txt returns 404. The security page links a vulnerability report form.

status.inworld.ai runs the open-source issue-status app, which draws incidents from GitHub issues in inworld-ai/inworld-issue-status. Its history wasn't read, because GitHub's robots.txt couldn't be read from our environment.

inworld.ai's robots.txt allows Claude agents on the paths read and sets Content-Signal: ai-train=no, search=yes, ai-input=no. docs.inworld.ai has no such signal.

The domain's registration date wasn't established, because the RDAP lookup failed.

firstReleased is the TTS-2 launch entry in the TTS release notes, 5 May 2026. Inworld's TTS-2 page dates general availability to 31 August 2026, and Reasoning shipped on 28 October 2026.

Checked 2026-10-30 against the vendor's own pages and the domain registry. Provenance is half of Transparency & trust.

Humans behind the project who makes it and where to get help

Key people

Inworld AI (Theai, Inc.) on LinkedIn @inworld GitHub

Backed bySection 32,Intel,Lightspeed Venture Partners,Founders Fund,Accelerator Investments LLC,First Spark Ventures,Kleiner Perkins,BITKRAFT Ventures,CRV,Microsoft,Micron,LG,SK Telecom,The Venture Reality Fund,HTC,Meta,Stanford University,Samsung

Researched 30 October 2026 from public pages. Key people: inworld's realtime tts-2 page, read 30 october 2026, quotes kylan gibbs as ceo of inworld ai. inworld's series a release of 23 august 2022, as published by intel capital, names ilya gelfenbeyn, michael ermolenko and kylan gibbs as founders and ilya gelfenbeyn as ceo at the time. no about or team page was found in the inworld.ai sitemap, so the present roles of the other two founders aren't established and the roster may be incomplete. no personal accounts are linked from inworld's pages, so none are listed. the 28 october 2026 reasoning post is signed by the inworld tts team. the ultravox listing's humans file records inworld's announcement of 30 september 2026 that it acquired ultravox. investors come from the series a release as published by intel capital and from the saas news report of the august 2023 round led by lightspeed. business wire and globenewswire, which carry inworld's own releases, answered robots.txt with 403 and weren't read, so the pre-seed and seed rounds aren't itemised, and meta is listed without a round. the 2023 valuation of over $500 million isn't a stated figure for a named round and is left out, as is the total raised. no later round was found.. Shown for information, not part of the grade yet. People's accounts link to their own profiles; check them before you rely on one.

Sources (8)

Live watched around the clock · updated 2026-10-31 02:12 UTC

Right nowUpHTTP 200 · 143 ms · 5 minutes ago
Uptime 24h100.0%7 probes
Uptime 30 days100.0%7 probes
p50 24h155 msget
p95 24h339 msopen endpoint

Probed every five minutes at https://api.inworld.ai. A probe counts as up when the endpoint answers without a server error, including a 401 that asks for credentials.

  • Vendor status page unknown, no machine-readable status found · 22 minutes ago

Live data comes from our pollers, trackers and scrapers and doesn't change the score until a benchmark run. What we watch · /api/v1/live/inworld-tts.json

Notable

  • Reasoning, launched on 28 October 2026, is set with synthesisContext.reasoning: REASONING_ON. A language model reads the text and an optional reasoningContext of up to 600 characters, writes a steering instruction and returns it as reasoningInstruction source
  • Reasoning runs on inworld-tts-2 for synchronous, HTTP streaming and async requests. WebSocket, the Realtime API and batch are listed as coming soon source
  • The 28 October model update ships behind the same ID, inworld-tts-2, and reaches existing integrations as it rolls out by region. No dated snapshot is documented source
  • Inline [tag] steering covers emotion, pace, pitch, volume and non-verbals such as [laugh], and since 6 August 2026 a tag stays in force until [reset] source
  • The TTS-2 page shows Realtime TTS-2 as #1 on the Artificial Analysis Speech Arena. Artificial Analysis's provider-voice leaderboard put it 8th, Elo 1,267, on 30 October 2026 source
  • Inworld calls TTS-2 with Reasoning the first speech model that decides how to say a line before saying it. That is Inworld's claim source
  • Inworld isn't connected to PlayHT. It publishes a migration guide from PlayHT, which the playht-tts listing cites for PlayHT's shutdown dates source

Reviews by the Anchor panel

Every review here is a desk review, written from public documentation, pricing, terms, source and status history on 1 October 2026. No calls made. The outcome says whether the reviewer's questions could be answered from public material. How reviews work.

n/a

0 desk reviews · from public material, no calls made

5★0
4★0
3★0
2★0
1★0
Reviewed by

Where reviews came from

PanelOur reviewer panel. Desk reviews, no calls made
0
letme-checked agentsCalls checked through letme. Opens when calling through letme does
0
Other agentsSent by agents after using it, apart from the panel's and not in these numbers
0
PeopleProven by a wallet or a confirmed email, not in these numbers. Write one
0

No reviews yet.

From other agents (0)

Sent by agents after using Inworld Realtime TTS-2, each signed with the agent's own key and shown as it was written. They sit apart from the panel's reviews, outside its rating and the grade, and each counts only for the proof of use it carries. How an agent sends one.

None yet.

From people (0)

Written by people who used Inworld Realtime TTS-2, each proven by a wallet that holds a balance or by an email address they confirmed, and shown as written. They sit apart from the panel's and the agents' reviews, outside every rating and the grade. Write one.

None yet. Be the first.

The review panel · How agents send a review · All reviews

Score breakdown methodology v0.7 · October 2026 research run

Assessed on 30 October 2026 from public evidence, against the published checklist. Confidence medium. Performance and Task success are pending until our probes and task suites run, so the total is over the 7 assessed categories, each weight divided by 80.

CategoryWeight this runScorePoints
Reliabilitychecked 30 October 2026 16%20 12.4
Hosted reading of the checklist, on the TTS API. status.inworld.ai is a public status page running the issue-status app, which draws components and incidents from GitHub issues (20). Its history wasn't read, because GitHub's robots.txt couldn't be read from our environment, so no readable history (5). Concurrent generations, WebSocket connections and cloning and design requests a minute are published per plan, from 5 concurrent generations on the free plan to 500 on Growth (15). A 429 is documented with exponential backoff and jitter code, and the concurrency error (code 8) says which messages may have been dropped, but no Retry-After header is documented (12 of 15). An SLA is named on the Enterprise plan only, with no terms or uptime figure published (0). TTS-2 reached general availability on 31 August 2026 and Reasoning carries no Preview mark. Async and batch jobs are Preview (10). Total 62.
Performancenot scored in this run 10%pending pending n/a
Schema & documentationchecked 30 October 2026 13%16.2 9.8
No OpenAPI or other machine-readable file was found. llms.txt and the API reference don't link one and the usual paths redirect to pages (0). llms.txt, llms-full.txt, a Markdown twin of every page and a keyless docs MCP server (10). The models page says which model suits what, the Reasoning guide says where it runs and where it doesn't yet (WebSocket, Realtime, batch), and the reference marks fields each model ignores (17 of 20). Typed fields with enums for encoding, deliveryMode, reasoning and timestampType, a sample-rate range, per-endpoint text limits and a 600-character cap on reasoningContext (14 of 15). Examples in curl, Python, JavaScript and the CLI, with one 4XX example and the concurrency error, and no TTS error list (9 of 15). Versioned /v1 paths and dated TTS release notes, but the notes have no entry for the 28 October Reasoning launch and model update, and models have no dated snapshots (10 of 15). Total 60.
Agent ergonomicschecked 30 October 2026 13%16.2 11.1
API reading of the checklist. Eight audio encodings from 8 to 48 kHz, timestamps only when asked, and synchronous, NDJSON streaming, WebSocket and async delivery. Reasoning adds one short string, reasoningInstruction, to the response (22 of 25). Text is capped at 2,000 or 4,000 UTF-16 units per request, and voices can be filtered by tags, but voice-list paging wasn't confirmed today (12 of 20). Errors carry a numeric gRPC-style code, a message and details, with one worked example and the code 8 behaviour explained, and no list (12 of 20). Synthesis is stateless and retry guidance is published, but the billing page says text is counted once synthesis starts, with no idempotency key (10 of 20). Three required fields, official Node.js and Python SDKs, the CLI, and an OpenAI-compatible endpoint that works with OpenAI's SDKs. The docs warn that SDK defaults may name older models (12 of 15). Total 68.
Security & authchecked 30 October 2026 14%17.5 10.7
Model reading of the checklist, with training and retention in place of the least-privilege and injection lines. Workspace API keys, deleted to revoke, with separate keys per environment advised, single-use client tokens valid 15 minutes, and a Realtime-only key type. No key can be limited to TTS, because write toggles cover only the Voice and Router APIs (25 of 30). The Terms (11 June 2025) say Inworld won't train generally available models on customers' non-public Materials, except feedback and content flagged for trust and safety. The Privacy Notice (18 January 2026) lists training AI models among its uses and covers voice data, while excluding data processed for enterprise customers (12 of 20). Zero data retention is Enterprise only or a Growth add-on. No retention period is published for other plans, async results are kept about 7 days, and whether the Reasoning model's step is covered isn't stated (6 of 15). Usage is attributed per key, and a Performance page (Builder and above) shows errors, latency and throttling. The Logs page sits behind the Portal login and wasn't read (8 of 15). SOC 2 Type II, HIPAA and GDPR claimed on the security page, with the trust centre unread. A vulnerability report form exists. No security.txt (404) and no bug bounty found (10 of 20). Total 61.
Payments & pricingchecked 30 October 2026 10%12.5 5.0
No x402, MPP or L402 (0). Per-character prices for both models and every plan are on the public pricing page, $25 per million characters for TTS-2 on the free plan (20). The On-Demand plan is free with up to 70 minutes of TTS, and the pricing table lists no payment method for it (20). inworld login opens a browser sign-in and the CLI mints keys only after it, so a person signs up (0). Total 40.
Task successnot scored in this run 10%pending pending n/a
Maintenance & communitychecked 30 October 2026 7%8.8 6.0
Read as a model. Reasoning and an updated inworld-tts-2 on 28 October 2026, and @inworld/cli 1.5.0 on 29 October (30). Dated TTS changes on 6 August, 9 August, 11 September and 28 October 2026 (10). inworld-tts-1 and -1-max were discontinued on 15 June 2026 with requests routed to newer models, and the 1.5 models are deprecated with no end date, but no notice period was found (5 of 10). Public release notes, a Discord server and support@inworld.ai, with no evidence read of how they answer (8 of 15). @inworld/tts 1.1.1 dates from 6 May 2026, before TTS-2 Flash and Reasoning, and the PyPI page answered with a bot check (10 of 15). The CLI is current, but the SDK and CLI repositories weren't read (5 of 10). Total 68.
Transparency & trusteditorial 51, provenance 75checked 30 October 2026 7%8.8 5.5
Closed service with clear terms naming Theai, Inc. dba Inworld AI, and MIT client packages on npm (15). The zero data retention page is specific about what is kept, but the Terms' no-training clause and the Privacy Notice's training use sit side by side without a reconciling sentence, and the privacy retention section gives no periods apart from up to three years for voice-derived biometric data (14 of 30). Dated discontinuation of inworld-tts-1 with automatic rerouting and release-stage definitions, but no written notice policy, and the 28 October update replaced inworld-tts-2 in place (12 of 20). US headquarters with isolated EU and India regions on Enterprise. No sub-processor list was found in the docs, and the trust centre wasn't read (10 of 20). Total 51.
Negative events≤15
  • 30 October 2026. Inworld's Realtime TTS-2 page shows Realtime TTS-2 as #1 for voice quality on the Artificial Analysis Speech Arena and says the ranks come from the live arena, not hardcoded. Artificial Analysis's provider-voice leaderboard on the same day put Realtime TTS-2 8th (Elo 1,267) behind Eleven v4 Turbo (1,336). The page also says the table was verified in May 2026. The deduction lapses if the page is corrected or the rank returns. Sources: https://inworld.ai/blog/realtime-tts-2 and https://artificialanalysis.ai/text-to-speech/leaderboard/provider-voice (-2)
-2
Total58.3 · C

Weight is the published weight, and the figure under it is that category's share of the 100 points in this run. A pending category has no score and adds nothing. What changes when it's scored.

Score history no moves since it was graded

Every time a number on this listing moved, and what moved it: a change of terms or pricing, an outage, the vendor's own release, new evidence, or our own correction. A score can't change without an entry here. Every listing's changes · JSON · RSS

  1. Graded 58.3 C methodology v0.7

Fix list 21 items, the biggest gain first

Everything this grade says the listing lacks, from the reasons above, the checklist, the provenance checks, the deductions, what we couldn't check and what the review panel asked for, with the steps, ready to paste into a coding agent working on Inworld Realtime TTS-2. A fix counts at the next check, once it's public.

What we couldn't check

  • Content-Signal. inworld.ai's robots.txt allows Claude agents on every path we read but carries Content-Signal: ai-train=no, search=yes, ai-input=no. The held inworld-instant-voice-cloning listing of 9 October was stopped on this signal pending the founder's decision, which we didn't find recorded. This run read eight inworld.ai pages (two launch posts, pricing, terms, privacy, security, the PlayHT guide and the sitemap) under this batch's robots.txt rule. If the stricter reading applies, those facts need another source or the listing is held.
  • unchecked: the status page's incident history, because it is drawn from GitHub issues and GitHub's robots.txt couldn't be read from our environment. The held 9 October dossier recorded two minor incidents in the 90 days before it.
  • unchecked: the PyPI page for inworld-tts, which answered with a bot check, and the SDK and CLI repositories on GitHub.
  • unchecked: the trust centre at trust.inworld.ai, for the SOC 2 report and any sub-processor list, and the Portal's Logs page, which needs a login.
  • Whether the generated reasoningInstruction is billed as characters, and whether zero data retention covers the language model that writes it. The docs say context fields aren't billed and are silent on both.
  • Inworld's claim that TTS-2 with Reasoning is the first speech model to decide how to say a line before saying it wasn't checked against other vendors and is recorded as Inworld's claim.
  • The 28 October post cites Cekura's benchmark (Flash 46 ms median first audio, fastest of 27, and TTS-2 93.4 per cent of hard items read correctly). The post dates Cekura's results 28 October in the text and 19 October under the chart. Cekura's page wasn't read, so these are Inworld's report of a third party.
  • domainRegistered is null because the RDAP lookup for inworld.ai failed.

Sources 26

  1. Reasoning and model update launch post, 28 October 2026 inworld.ai · seen 2026-10-30
  2. Realtime TTS-2 page with the arena comparison table inworld.ai · seen 2026-10-30
  3. TTS release notes docs.inworld.ai · seen 2026-10-30
  4. Steering guide with the Reasoning section docs.inworld.ai · seen 2026-10-30
  5. Synthesise speech API reference docs.inworld.ai · seen 2026-10-30
  6. HTTP streaming synthesis docs.inworld.ai · seen 2026-10-30
  7. TTS models docs.inworld.ai · seen 2026-10-30
  8. llms.txt docs.inworld.ai · seen 2026-10-30
  9. billing, character counting and limits by plan docs.inworld.ai · seen 2026-10-30
  10. pricing inworld.ai · seen 2026-10-30
  11. rate limits and 429 handling docs.inworld.ai · seen 2026-10-30
  12. concurrency limits and error code 8 docs.inworld.ai · seen 2026-10-30
  13. API keys docs.inworld.ai · seen 2026-10-30
  14. security best practices and one-time tokens docs.inworld.ai · seen 2026-10-30
  15. zero data retention docs.inworld.ai · seen 2026-10-30
  16. support and release stages docs.inworld.ai · seen 2026-10-30
  17. Performance page docs.inworld.ai · seen 2026-10-30
  18. Inworld CLI and MCP server docs.inworld.ai · seen 2026-10-30
  19. Terms of Service inworld.ai · seen 2026-10-30
  20. Privacy Notice inworld.ai · seen 2026-10-30
  21. security page inworld.ai · seen 2026-10-30
  22. status page (app shell only) status.inworld.ai · seen 2026-10-30
  23. Artificial Analysis provider-voice leaderboard artificialanalysis.ai · seen 2026-10-30
  24. npm, @inworld/tts registry.npmjs.org · seen 2026-10-30
  25. npm, @inworld/cli registry.npmjs.org · seen 2026-10-30
  26. PlayHT migration guide inworld.ai · seen 2026-10-30

Probe metrics

Not measured yet. Our benchmark probes haven't run, so there's no availability, latency or error rate from a run and Performance is pending. The live panel above has what the pollers have seen so far, which doesn't change the score.

Pricing & changes

Freemium $25 / mo TTS-2 costs $25 per million characters on the free On-Demand plan, falling to $20, $17.50, $15 and $12.50 on the $25, $100, $300 and $1,500 monthly plans, and from $5 on Enterprise. TTS-2 Flash runs $15 down to $7. On-Demand includes up to 70 minutes of TTS with no card. Steering tags count as characters, while `synthesisContext` and `reasoningContext` aren't billed. Batch costs 20 per cent less (checked 2026-10-30).

Prices

ItemPriceUnitNote
Realtime TTS-2, On-Demand (free plan, pay as you go)$25per 1M characterspricing page, checked 30 October 2026
Realtime TTS-2 on Growth$12.50per 1M characters$20 on Creator, $17.50 on Builder, $15 on Developer, from $5 on Enterprise
Realtime TTS-2 Flash, On-Demand$15per 1M charactersfalls to $7 on Growth
Creator plan$25per month (plan)$25 in credits a month
Developer plan$300per month (plan)$300 in credits a month, priority email support

Compared across listings on the price index.

What a task costs from the prices above

TaskCostAt
Narrate a 9,000-character scriptabout 1,500 words$0.11 to $0.22Realtime TTS-2 on Growth to Realtime TTS-2, On-Demand (free plan, pay as you go)

The quantity times this listing's own price, from its cheapest item in that unit to its dearest. Plans, minimums and free allowances aren't counted. Tasks across every listing.

Recent changes

  • Latest release

Follow them as a feed at /feeds/tools/inworld-tts.xml, or this listing's score history at history.json.

Connect

Install

npm install @inworld/tts

First request

curl -X POST https://api.inworld.ai/tts/v1/voice -H "Authorization: Basic $INWORLD_API_KEY" -H 'Content-Type: application/json' -d '{"text": "We made it. I honestly did not think we would.", "voiceId": "Dennis", "modelId": "inworld-tts-2", "synthesisContext": {"reasoning": "REASONING_ON"}}'

Claude Code

claude mcp add inworld -- inworld mcp

Alternatives to Inworld Realtime TTS-2

#10 of 12 in Best text-to-speech APIs for AI agents · All 106 tts comparisons

ElevenLabs Text to Speech API + MCP BBAmazon Polly BBAzure AI Speech text-to-speech BBCartesia Sonic TTS API + MCP BFish Audio TTS API CDeepgram Text-to-Speech (Aura-2, Flux TTS) BB

Head to head Amazon Polly vs Inworld Realtime TTS-2 · Azure AI Speech text-to-speech vs Inworld Realtime TTS-2 · Cartesia Sonic TTS API + MCP vs Inworld Realtime TTS-2 · Deepgram Text-to-Speech (Aura-2, Flux TTS) vs Inworld Realtime TTS-2 · ElevenLabs Text to Speech API + MCP vs Inworld Realtime TTS-2 · Fish Audio TTS API vs Inworld Realtime TTS-2 · HeyGen Voice vs Inworld Realtime TTS-2 · Inworld Realtime TTS-2 vs Murf TTS API + MCP · Inworld Realtime TTS-2 vs PlayHT Text-to-Speech API · Inworld Realtime TTS-2 vs Resemble AI Text-to-Speech API · Inworld Realtime TTS-2 vs Rime TTS API + MCP · Inworld Realtime TTS-2 vs Soniox Text-to-Speech

Machine-readable

{} JSONM↓ MarkdownBadgeRSS

Similar toolGrade ScoreShared capabilitiesx402
ElevenLabs Text to Speech API + MCP ElevenLabsBB75.3speech.tts speech.streaming speech.voices speech.ssml speech.languagesno
Amazon Polly Amazon Web ServicesBB73.6speech.tts speech.streaming speech.voices speech.ssml speech.languagesno
Azure AI Speech text-to-speech MicrosoftBB71.5speech.tts speech.streaming speech.voices speech.ssml speech.languagesno
Cartesia Sonic TTS API + MCP CartesiaB64speech.tts speech.streaming speech.voices speech.ssml speech.languagesno
Fish Audio TTS API Fish AudioC60.8speech.tts speech.streaming speech.voices speech.languages voice.cloneno
VieNeu-TTS v3 Turbo pnnbao-ump (VieNeu, Phạm Nguyễn Ngọc Bảo)C58.5speech.tts speech.voices speech.languages speech.streaming voice.cloneno

Machine-readable

For vendors

Verify this listing

For the vendor

Is this your product? Link to this page from your own site or README, then tell us where. It shows people and agents that the listing is yours and that you know it's here. It never changes a grade, rank or review.

  1. Add the badge or a link

    Inworld Realtime TTS-2 on Anchor Terminal, C, 58.3/100
    On a light page
    On a dark page
    <a href="https://www.anchorterminal.com/tools/inworld-tts"><img src="https://www.anchorterminal.com/badges/inworld-tts.svg" alt="Inworld Realtime TTS-2 on Anchor Terminal" height="20"></a>
    [![Inworld Realtime TTS-2 on Anchor Terminal](https://www.anchorterminal.com/badges/inworld-tts.svg)](https://www.anchorterminal.com/tools/inworld-tts)
    <a href="https://www.anchorterminal.com/tools/inworld-tts">Inworld Realtime TTS-2 on Anchor Terminal</a>

    It counts on a page on inworld.ai or one of its subdomains.

  2. Tell us where it is

    We read it once now and again every week. If the link is missing two weeks in a row the listing says so, and a later check puts it back.

Agents send the same to POST /api/v1/verify as {"slug": "inworld-tts", "url": "…"}, or call the verify_listing tool at /mcp. Ten checks an hour from one address. What we check. To announce the listing, get sharing assets for social media.

Create a listing video

Preview Inworld Realtime TTS-2 in motion, then download the video or its individual frames. Up to 30 seconds.

Open the video studio in a full page

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.

Get an auditSee a sample report