Inworld Realtime API vs Rime Prism

Inworld Realtime API 60

No AsyncAPI or other machine-readable file for the socket, and the spec paths redirect (0). llms.txt, Markdown twins and a docs MCP server (10). The overview, OpenAI migration guide, providerData reference and the billing page's account of how a session is counted say what to use when (16 of 20). The reference lists every client and server event with field types and required marks, as text rather than a schema (12 of 15). JavaScript and Python example projects and an error event with type, code and param, without a list of codes (9 of 15). Dated release notes with affected clients and required actions, but no API version beyond /v1 (13 of 15). Total 60.

Rime Prism 69

No OpenAPI or AsyncAPI file for the WebSocket API was found in llms.txt or the reference (0 of 25). llms.txt and a Markdown twin of every page (10). The overview says what Prism handles, what it doesn't (text-only replies, sampling settings, a choice of language model, ending the caller's turn) and when a pipeline fits better, and a chart says when to use the SDK, LiveKit, Pipecat or the raw WebSocket (18 of 20). The WebSocket reference types every client and server event field with defaults and allowed values, refuses unknown fields with invalid_client_event naming the field, and takes tool parameters as JSON Schema. The schemas exist only as tables, not a file (13 of 15). Python and Node.js examples on each guide, and an errors page that lists 41 codes with cause, scope and what to do, plus HTTP refusals and close codes (15). A dated changelog with an RSS feed, a version field in session.created (such as 20261023a) and pinned SDK versions. The path carries /v1, with no version policy for the event set found. Partly met, so the earlier 13 is kept (13 of 15). Docs and changelog re-read on 4 November. 0+10+18+13+15+13 = 69.

Inworld Realtime API 63

Closed service with clear terms naming Theai, Inc. dba Inworld AI (15). The ZDR page says what a session keeps and that third-party LLM requests aren't covered, but the terms and privacy notice disagree on training (12 of 30). Release notes flag potentially breaking changes with the clients affected and what to do, without stating a notice period (14 of 20). US headquarters with EU and India residency on Enterprise, and no sub-processor list in the docs, while sessions may route to third-party LLM providers (10 of 20). Total 51.

Rime Prism 72

Closed service with clear terms from Rime Labs, Inc., and the SDK is MIT (15 of 30). The terms changed on 29 October 2026. Section 6.3(b) now grants Rime a perpetual licence to train and improve its models on Free Plan customer data (excluding voice recordings, audio with a natural person's voice and third parties' personal data), while the privacy policy (effective 22 August 2025) says "We do not use customer data to train AI models unless explicitly requested and agreed upon" and the security page says "We never use customer text or audio to train our models". The privacy and security pages still describe text in and audio out, and none covers Prism's caller audio, while the new terms define Call Content and say not to send Caller audio on a Free Plan. The DPA at rime.ai/dpa couldn't be reached from our network on 28 October. The contradiction between the terms and the security page is charged once, as the misleading-claim deduction, so the line holds (15 of 30). The new terms promise at least 30 days' notice before retiring a Model Version used on a paid plan. Arcana's cloud retirement came with 17 days' notice before that policy (15 of 20, was 10). A sub-processor list with locations, all in the United States, and the Prism endpoint stated as US East (us-east-2), per the 28 October check (18 of 20). 15+15+15+18 = 63.

Inworld Realtime API 62

Hosted reading of the checklist. status.inworld.ai is a public status page (20), with history we couldn't read, as on inworld-tts (5). Concurrent sessions are published per plan, 10 on On-Demand up to 1,000 on Growth (15). The rate limits page gives exponential backoff with jitter and Retry-After on longer windows, and sessions emit rate_limits.updated, with nothing on reconnecting a dropped session (12 of 15). An SLA appears only in the Enterprise column, with no terms (0). No preview or beta label on the Realtime API (10). 20 + 5 + 15 + 12 + 0 + 10 = 62.

Rime Prism 55

status.rime.ai redirects to an incident.io page at rimestat.us with Text-to-speech API and Platform API components, and still no Prism component on 4 November 2026. A vendor status page with no component for the product takes 15 in this run (15, was 10). Its feed lists two incidents, Text-to-speech partial degradations on 3 and 5 September, and nothing since Prism launched on 28 October, a short record of seven days (15, was 5). The errors page publishes per-session numbers (55 seconds a turn, 60 seconds of queued speech, 120 seconds or 512 tokens a reply, 5 minutes idle, 1 MiB a message), but no concurrency limit per account or key, and a full server answers 503. Partly met, so the earlier 8 is kept (8). Overload is documented as capacity_exceeded, unavailable, a 503 refusal ("Retry after a backoff") and close code 1001, the page says which reply failures can succeed when sent again, and the tools guide says never to rerun a call with an unknown outcome. Safe-retry guidance without idempotency keys gives 12 in this run (12, was 13, not a checklist value). SLAs are an Enterprise item and none is published (0). The changelog calls Prism available, but all four SDKs are 0.1.0 alpha releases, the Pipecat path uses a pinned preview, and the pricing page says half duplex at launch. Partly met, so the earlier 5 is kept (5). 15+15+8+12+0+5 = 55.

Inworld Realtime API 68

Scored as an API. Conversation history defaults to the last 100 items, can be trimmed, and can be cached on Anthropic models, and every response.done carries a usage block (18 of 25). No lists to page. Output controls are modalities, turn detection settings, optional transcription and response.speak (12 of 20). Error events follow OpenAI's type and code shape, unsupported fields come back as non-fatal unsupported_field warnings, and item_not_found names the bad parameter (14 of 20). Tools run on the client and a follow-up response starts automatically after a tool result. Nothing covers side effects of a tool call cut off by an interruption (12 of 20). Works with OpenAI Realtime clients after changing the URL and key, with no Inworld Realtime SDK (12 of 15). Total 68.

Rime Prism 73

Scored as a model API. A 65,536-token context, with truncation set to leave out the oldest turns by default, a configurable retention_ratio, a prism_truncation report on each reply, prism.conversation.get to read the stored history and conversation.item.delete to trim it (20 of 25). Replies are capped at 120 seconds or 512 tokens and can't be set lower. max_response_output_tokens, temperature and text-only replies are refused. Up to 16 labels per reply for routing. No lists to page. Partly met, so the earlier 8 is kept (8 of 20). Errors carry code, scope, owner, param and a correlation_id, and the docs tell clients to branch on code and scope, never on message (20). Client events are matched by event_id, a repeated identical tool result is acknowledged, and Prism records late results. Prism doesn't deduplicate requests, a dropped connection doesn't resume, and the client replays history into a new session. Partly met, so the earlier 12 is kept (12 of 20). Short instructions are enough to start, the voice defaults to astra and interruption is on. Official SDKs in Python, Node.js, Go and Rust, all alpha, plus LiveKit and Pipecat quickstarts. Partly met while the SDKs are alpha, so the earlier 13 is kept (13 of 15). 20+8+20+12+13 = 73.