{ "data": { "similar": [ { "grade": "B", "json": "https://www.anchorterminal.com/tools/retell-ai.json", "name": "Retell AI API + MCP", "score": 69.1, "shared": [ "voice.agent", "voice.speech-to-speech", "voice.tools", "voice.telephony" ], "slug": "retell-ai" }, { "grade": "B", "json": "https://www.anchorterminal.com/tools/vapi.json", "name": "Vapi API + MCP", "score": 63.5, "shared": [ "voice.agent", "voice.speech-to-speech", "voice.tools", "voice.telephony" ], "slug": "vapi" }, { "grade": "C", "json": "https://www.anchorterminal.com/tools/ultravox.json", "name": "Ultravox Realtime API", "score": 58.5, "shared": [ "voice.agent", "voice.speech-to-speech", "voice.tools", "voice.telephony" ], "slug": "ultravox" }, { "grade": "C", "json": "https://www.anchorterminal.com/tools/hume-evi.json", "name": "Hume EVI (Empathic Voice Interface)", "score": 56.7, "shared": [ "voice.agent", "voice.speech-to-speech", "voice.tools", "voice.telephony" ], "slug": "hume-evi" }, { "grade": "D", "json": "https://www.anchorterminal.com/tools/bolna.json", "name": "Bolna API + MCP", "score": 52.6, "shared": [ "voice.agent", "voice.speech-to-speech", "voice.tools", "voice.telephony" ], "slug": "bolna" }, { "grade": "BB", "json": "https://www.anchorterminal.com/tools/elevenlabs-agents.json", "name": "ElevenLabs Agents API + MCP", "score": 71.3, "shared": [ "voice.agent", "voice.tools", "voice.telephony" ], "slug": "elevenlabs-agents" } ], "tool": { "slug": "openai-realtime", "name": "OpenAI Realtime API", "vendor": "OpenAI", "vendorUrl": "https://openai.com", "kind": "http-api", "category": "voice-agents", "summary": "OpenAI's Realtime API runs spoken conversations with speech-to-speech models such as `gpt-realtime-2.1`. Clients connect over WebRTC, WebSocket or SIP, and sessions support interruptions, function calling and remote MCP tools.", "url": "https://www.anchorterminal.com/tools/openai-realtime", "markdownUrl": "https://www.anchorterminal.com/tools/openai-realtime.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/openai-realtime.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/openai-realtime.json", "repo": "https://github.com/openai/openai-agents-js", "license": "Proprietary service. The service terms on openai.com were not read. The Agents SDK for TypeScript and the OpenAPI description are MIT", "transports": [ "http" ], "remoteUrl": "https://api.openai.com/v1/realtime", "packages": [ { "registry": "npm", "name": "@openai/agents" }, { "registry": "npm", "name": "@openai/agents-realtime" } ], "auth": "api-key", "authNotes": "Self-serve API key from platform.openai.com, sent as a Bearer header on server WebSocket connections and call-control requests. For browsers and phones, a server calls `POST /v1/realtime/client_secrets` for a client secret (`ek_`) that lasts 10 seconds to 2 hours, 10 minutes by default, and can start several sessions until it expires. The browser WebSocket example passes the secret as a subprotocol. SIP calls are tied to a project by the project ID in the SIP address.", "pricing": "usage", "pricingNotes": "Per 1M tokens on `gpt-realtime-2.1`. Audio $32.00 in, $0.40 cached and $64.00 out, text $4.00 in and $24.00 out, image $5.00 in. `gpt-realtime-2.1-mini` is $10.00 and $20.00 for audio. Each response re-sends the whole conversation, and input transcription is billed separately. Usage tiers start at $5 of credit purchases. Whether the Free tier covers Realtime models was not established (https://developers.openai.com/api/docs/pricing.md, checked 2026-10-09).", "priceSummary": "Pay per use", "where": "hosted", "x402": { "level": "no", "evidence": "No x402, MPP or L402 in the Realtime guides, the pricing page or the OpenAPI file (checked 2026-10-09).", "endpoints": [] }, "toolCount": null, "popularity": { "githubStars": null, "npmWeekly": null, "pypiWeekly": null, "asOf": "2026-10-09" }, "docsUrl": "https://developers.openai.com/api/docs/guides/realtime", "llmsTxt": "https://developers.openai.com/api/docs/llms.txt", "openapi": "https://github.com/openai/openai-openapi/blob/main/openapi.yaml", "capabilities": [ "voice.agent", "voice.speech-to-speech", "voice.tools", "voice.telephony" ], "tags": [ "hosted", "usage-priced", "closed-source", "speech-to-speech", "webrtc", "websocket", "sip", "streaming", "typescript", "python", "openapi", "llms-txt", "api-key", "function-calling", "mcp-client", "status-page" ], "lastRelease": "2026-07-28", "graded": true, "anchor": { "graded": true, "score": 63.3, "grade": "B", "agentReady": false, "rank": 400, "ranked": true, "rankOf": 950, "categoryRank": 7, "methodology": "0.4", "run": "2026-10-01", "scores": { "ergonomics": 76, "maintenance": 52, "payments": 20, "reliability": 67, "schema": 84, "security": 61, "transparency": 71 }, "pending": [ "performance", "tasks" ], "breakdown": [ { "key": "reliability", "name": "Reliability", "weight": 16, "effectiveWeight": 20, "score": 67, "points": 13.4, "reason": "Graded as a hosted API. status.openai.com lists Realtime as one of 13 API components with history (20). The page shows 100 per cent uptime for the Realtime component for July to October 2026, and the status feed names Realtime among the affected components of five resolved incidents on 24 July, 25 July (two), 17 September and 6 October. The feed gives only the final update, so durations were not read, and we score them as minor (20 of 30). The rate limits page gives usage tiers and monthly usage limits, but per-model limits for Realtime are shown only in account settings and the public table covers GPT-6 families only (5 of 15). The docs describe `Retry-After` on 429 and 503, exponential backoff with jitter and SDK retries, sessions emit `rate_limits.updated`, and SIP webhooks carry a `webhook-id` for deduplication. No idempotency key for call control was found (12 of 15). No SLA was found in the pages read, and openai.com answered 403 (0). The Realtime API has been generally available since 28 August 2025 (10)." }, { "key": "performance", "name": "Performance", "weight": 10, "effectiveWeight": 0, "pending": true, "points": 0, "reason": "Pending. Latency is measured per call by our probes, which haven't run yet, so this run doesn't score it. Its weight is shared across the assessed categories until the first probe window closes." }, { "key": "schema", "name": "Schema \u0026 documentation", "weight": 13, "effectiveWeight": 16.25, "score": 84, "points": 13.65, "reason": "The OpenAPI 3.1 file in `openai/openai-openapi` (updated 8 October 2026) has nine `/realtime` paths and 167 Realtime schemas, with 13 client events and 48 server events typed. It is not an AsyncAPI document, so the WebSocket channel itself is not described (22 of 25). llms.txt and a Markdown twin of every docs page (10). The tools guide has a table of when to use `function`, `mcp` with `server_url` or a connector, and the transport guides say when to choose WebRTC, WebSocket or SIP (16 of 20). Session fields carry enums and ranges, such as `expires_after.seconds` from 10 to 7200, though `model` also accepts any string (13 of 15). Examples in JavaScript, Python, Java, Ruby and C#, an `error` event with type, code, message and param, and a list of common MCP failures. No catalogue of Realtime error codes was found (11 of 15). Dated changelog and dated model snapshots, less 3 because the OpenAPI file still lists `gpt-4o-realtime-preview` models and beta event schemas that the deprecations page says were removed in May 2026, and several guides mix GPT-Live and Realtime text on one page (12 of 15)." }, { "key": "ergonomics", "name": "Agent ergonomics", "weight": 13, "effectiveWeight": 16.25, "score": 76, "points": 12.35, "reason": "Scored as a model API. Context can be capped with `truncation`, `retention_ratio` and `token_limits.post_instructions`, items can be deleted, prompt caching is automatic, and `response.done` reports token usage by modality (22 of 25). There are no lists to page. Output controls are `output_modalities`, `max_output_tokens`, optional input transcription and `allowed_tools` (14 of 20). The `error` event returns the `event_id` of the client event that caused it, and MCP tool listing and calls have their own failure events (15 of 20). Client events take an `event_id`, webhooks carry a `webhook-id`, and interruptions cancel the response. No session resumption after a dropped connection and no idempotency key for call control were found in the pages read (10 of 20). A session needs only a model, voice activity detection is automatic, and the Agents SDK starts a browser voice agent in about ten lines, with examples in five languages (15)." }, { "key": "security", "name": "Security \u0026 auth", "weight": 14, "effectiveWeight": 17.5, "score": 61, "points": 10.68, "reason": "Server connections send the API key as a Bearer header. Browsers use client secrets from `POST /v1/realtime/client_secrets` that last 10 seconds to 2 hours, 10 minutes by default, and can open several sessions until expiry. The browser WebSocket example passes the secret as a subprotocol, not in the query string. The docs index also lists workload identity federation, mutual TLS and an IP allowlist. Key scopes were not read (24 of 30). MCP tools take `allowed_tools` and `require_approval`, which defaults to `always` in the OpenAPI file, and a sideband connection keeps tool execution on the server. Session settings attached to a client secret can be overridden by the client connection (15 of 20). The model hears callers and reads MCP results. The SIP guide says to treat SIP headers as untrusted, and no prompt-injection guidance specific to Realtime was found (6 of 15). The docs index describes Admin APIs with audit log retrieval, sessions can write traces, and usage is reported per response (10 of 15). openai.com answered 403, so security.txt, any bug bounty and certifications are unchecked. We read a private vulnerability reporting policy on the OpenAPI repository and a 5 October changelog entry on HIPAA support (6 of 20)." }, { "key": "payments", "name": "Payments \u0026 pricing", "weight": 10, "effectiveWeight": 12.5, "score": 20, "points": 2.5, "reason": "No x402, MPP or L402 in the docs or the OpenAPI file (0 of 40). Per-token prices for every Realtime model are published without a login (20). The rate limits page lists a Free tier with a $100 monthly usage limit, but whether it covers Realtime models without a card was not established, and the error guide ties access to prepaid credits (0 of 20). API keys are created in the platform settings after a browser sign-in (0)." }, { "key": "tasks", "name": "Task success", "weight": 10, "effectiveWeight": 0, "pending": true, "points": 0, "reason": "Pending. Task success needs the category task suites run through each tool, which haven't run yet, so this run doesn't score it. Its weight is shared across the assessed categories until then. A data provider's data-quality score is published on its listing now and becomes half of this category when it's scored." }, { "key": "maintenance", "name": "Maintenance \u0026 community", "weight": 7, "effectiveWeight": 8.75, "score": 52, "points": 4.55, "reason": "The last dated changelog entry tagged `v1/realtime` that adds a feature is 28 July 2026, 73 days before the check, when `gpt-transcribe` became available for final transcripts of Realtime turns. `gpt-realtime-2.1` shipped on 6 July (20 of 30). Two entries tagged `v1/realtime` fall in the last 90 days, 28 July and 26 August, one short of the three the checklist asks for (0 of 20). Closed service with a changelog updated several times a week, plus a community forum and Discord linked from the docs, which we did not test (10 of 15). Current official SDKs, with `@openai/agents-realtime` 0.20.0 tagged on 8 October 2026 and 12 version tags since 29 July (15). The SDK is MIT but still 0.x, and its CI results were not read (7 of 10)." }, { "key": "transparency", "name": "Transparency \u0026 trust", "weight": 7, "effectiveWeight": 8.75, "score": 71, "points": 6.21, "note": "editorial 58, provenance 84", "reason": "Closed service. The SDKs and the OpenAPI file are MIT. The service terms were not read because openai.com answered 403 (10 of 30). The data controls page says API data is not used for training unless the customer opts in, abuse monitoring logs are kept up to 30 days, `/v1/realtime` stores no application state and is eligible for Zero Data Retention with approval. The privacy policy and any DPA were not read, so agreement between them is unchecked (18 of 30). The deprecations page states at least six months' notice for generally available models and lists shutdown dates with replacements (20 of 20). Data residency exists for the United States and Europe, with a `sip-eu.api.openai.com` endpoint, and tracing on `/v1/realtime` is not EU residency compliant. The sub-processor list is linked but was not read (10 of 20)." } ], "assessment": { "date": "2026-10-09", "basis": "public evidence", "confidence": "low", "notes": { "ergonomics": "Scored as a model API. Context can be capped with `truncation`, `retention_ratio` and `token_limits.post_instructions`, items can be deleted, prompt caching is automatic, and `response.done` reports token usage by modality (22 of 25). There are no lists to page. Output controls are `output_modalities`, `max_output_tokens`, optional input transcription and `allowed_tools` (14 of 20). The `error` event returns the `event_id` of the client event that caused it, and MCP tool listing and calls have their own failure events (15 of 20). Client events take an `event_id`, webhooks carry a `webhook-id`, and interruptions cancel the response. No session resumption after a dropped connection and no idempotency key for call control were found in the pages read (10 of 20). A session needs only a model, voice activity detection is automatic, and the Agents SDK starts a browser voice agent in about ten lines, with examples in five languages (15).", "maintenance": "The last dated changelog entry tagged `v1/realtime` that adds a feature is 28 July 2026, 73 days before the check, when `gpt-transcribe` became available for final transcripts of Realtime turns. `gpt-realtime-2.1` shipped on 6 July (20 of 30). Two entries tagged `v1/realtime` fall in the last 90 days, 28 July and 26 August, one short of the three the checklist asks for (0 of 20). Closed service with a changelog updated several times a week, plus a community forum and Discord linked from the docs, which we did not test (10 of 15). Current official SDKs, with `@openai/agents-realtime` 0.20.0 tagged on 8 October 2026 and 12 version tags since 29 July (15). The SDK is MIT but still 0.x, and its CI results were not read (7 of 10).", "payments": "No x402, MPP or L402 in the docs or the OpenAPI file (0 of 40). Per-token prices for every Realtime model are published without a login (20). The rate limits page lists a Free tier with a $100 monthly usage limit, but whether it covers Realtime models without a card was not established, and the error guide ties access to prepaid credits (0 of 20). API keys are created in the platform settings after a browser sign-in (0).", "reliability": "Graded as a hosted API. status.openai.com lists Realtime as one of 13 API components with history (20). The page shows 100 per cent uptime for the Realtime component for July to October 2026, and the status feed names Realtime among the affected components of five resolved incidents on 24 July, 25 July (two), 17 September and 6 October. The feed gives only the final update, so durations were not read, and we score them as minor (20 of 30). The rate limits page gives usage tiers and monthly usage limits, but per-model limits for Realtime are shown only in account settings and the public table covers GPT-6 families only (5 of 15). The docs describe `Retry-After` on 429 and 503, exponential backoff with jitter and SDK retries, sessions emit `rate_limits.updated`, and SIP webhooks carry a `webhook-id` for deduplication. No idempotency key for call control was found (12 of 15). No SLA was found in the pages read, and openai.com answered 403 (0). The Realtime API has been generally available since 28 August 2025 (10).", "schema": "The OpenAPI 3.1 file in `openai/openai-openapi` (updated 8 October 2026) has nine `/realtime` paths and 167 Realtime schemas, with 13 client events and 48 server events typed. It is not an AsyncAPI document, so the WebSocket channel itself is not described (22 of 25). llms.txt and a Markdown twin of every docs page (10). The tools guide has a table of when to use `function`, `mcp` with `server_url` or a connector, and the transport guides say when to choose WebRTC, WebSocket or SIP (16 of 20). Session fields carry enums and ranges, such as `expires_after.seconds` from 10 to 7200, though `model` also accepts any string (13 of 15). Examples in JavaScript, Python, Java, Ruby and C#, an `error` event with type, code, message and param, and a list of common MCP failures. No catalogue of Realtime error codes was found (11 of 15). Dated changelog and dated model snapshots, less 3 because the OpenAPI file still lists `gpt-4o-realtime-preview` models and beta event schemas that the deprecations page says were removed in May 2026, and several guides mix GPT-Live and Realtime text on one page (12 of 15).", "security": "Server connections send the API key as a Bearer header. Browsers use client secrets from `POST /v1/realtime/client_secrets` that last 10 seconds to 2 hours, 10 minutes by default, and can open several sessions until expiry. The browser WebSocket example passes the secret as a subprotocol, not in the query string. The docs index also lists workload identity federation, mutual TLS and an IP allowlist. Key scopes were not read (24 of 30). MCP tools take `allowed_tools` and `require_approval`, which defaults to `always` in the OpenAPI file, and a sideband connection keeps tool execution on the server. Session settings attached to a client secret can be overridden by the client connection (15 of 20). The model hears callers and reads MCP results. The SIP guide says to treat SIP headers as untrusted, and no prompt-injection guidance specific to Realtime was found (6 of 15). The docs index describes Admin APIs with audit log retrieval, sessions can write traces, and usage is reported per response (10 of 15). openai.com answered 403, so security.txt, any bug bounty and certifications are unchecked. We read a private vulnerability reporting policy on the OpenAPI repository and a 5 October changelog entry on HIPAA support (6 of 20).", "transparency": "Closed service. The SDKs and the OpenAPI file are MIT. The service terms were not read because openai.com answered 403 (10 of 30). The data controls page says API data is not used for training unless the customer opts in, abuse monitoring logs are kept up to 30 days, `/v1/realtime` stores no application state and is eligible for Zero Data Retention with approval. The privacy policy and any DPA were not read, so agreement between them is unchecked (18 of 30). The deprecations page states at least six months' notice for generally available models and lists shutdown dates with replacements (20 of 20). Data residency exists for the United States and Europe, with a `sip-eu.api.openai.com` endpoint, and tracing on `/v1/realtime` is not EU residency compliant. The sub-processor list is linked but was not read (10 of 20)." }, "sources": [ { "what": "Getting started with the Realtime API (Markdown twin)", "url": "https://developers.openai.com/api/docs/guides/realtime.md", "seen": "2026-10-09" }, { "what": "Realtime with tools, function and MCP tools, approvals", "url": "https://developers.openai.com/api/docs/guides/realtime-mcp.md", "seen": "2026-10-09" }, { "what": "Realtime conversations, events, errors, interruption", "url": "https://developers.openai.com/api/docs/guides/realtime-conversations.md", "seen": "2026-10-09" }, { "what": "WebSockets guide, authentication for server and browser", "url": "https://developers.openai.com/api/docs/guides/voice-websockets.md", "seen": "2026-10-09" }, { "what": "Telephony and SIP guide", "url": "https://developers.openai.com/api/docs/guides/voice-sip.md", "seen": "2026-10-09" }, { "what": "Server-side controls, sideband connections", "url": "https://developers.openai.com/api/docs/guides/voice-server-controls.md", "seen": "2026-10-09" }, { "what": "Cost guide, Realtime token accounting, caching and truncation", "url": "https://developers.openai.com/api/docs/guides/voice-latency-cost.md", "seen": "2026-10-09" }, { "what": "Migrate to GPT-Live guide", "url": "https://developers.openai.com/api/docs/guides/live-migration.md", "seen": "2026-10-09" }, { "what": "Pricing", "url": "https://developers.openai.com/api/docs/pricing.md", "seen": "2026-10-09" }, { "what": "Rate limits and usage tiers", "url": "https://developers.openai.com/api/docs/guides/rate-limits.md", "seen": "2026-10-09" }, { "what": "Error codes", "url": "https://developers.openai.com/api/docs/guides/error-codes.md", "seen": "2026-10-09" }, { "what": "Deprecations and notice periods", "url": "https://developers.openai.com/api/docs/deprecations.md", "seen": "2026-10-09" }, { "what": "API changelog", "url": "https://developers.openai.com/api/docs/changelog.md", "seen": "2026-10-09" }, { "what": "Data controls, retention and residency", "url": "https://developers.openai.com/api/docs/guides/your-data.md", "seen": "2026-10-09" }, { "what": "Create client secret reference", "url": "https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create.md", "seen": "2026-10-09" }, { "what": "Docs index (llms.txt) for the API guides", "url": "https://developers.openai.com/api/docs/llms.txt", "seen": "2026-10-09" }, { "what": "OpenAPI 3.1 description, read as a file from the repository", "url": "https://github.com/openai/openai-openapi", "seen": "2026-10-09" }, { "what": "Agents SDK for TypeScript, tags and `@openai/agents-realtime` manifest, read from a clone", "url": "https://github.com/openai/openai-agents-js", "seen": "2026-10-09" }, { "what": "Status page, components and uptime", "url": "https://status.openai.com/", "seen": "2026-10-09" }, { "what": "Status feed, incidents since July 2026", "url": "https://status.openai.com/feed.atom", "seen": "2026-10-09" }, { "what": "RDAP record for openai.com", "url": "https://rdap.verisign.com/com/v1/domain/openai.com", "seen": "2026-10-09" }, { "what": "robots.txt for developers.openai.com (200, allows all), openai.com (200, allows all but one path) and status.openai.com (404, so no rules)", "url": "https://developers.openai.com/robots.txt", "seen": "2026-10-09" } ], "openQuestions": [ "unchecked: the service terms, privacy policy, DPA, sub-processor list, security and trust pages and security.txt on openai.com. The host answered 403 to one request on 9 October 2026 and we did not retry. `provenance.terms` and `provenance.privacy` are left out for that reason.", "unchecked: whether an SLA is published for any tier. None was found in the developer docs. The rate limits page links Scale Tier and Reserved Tier pages on openai.com, which were not read.", "unchecked: whether the Free usage tier gives access to Realtime models without a card or prepaid credits.", "unchecked: duration and severity of the five incidents that name Realtime. The status feed shows only the final update of each.", "unchecked: per-model rate limits and concurrent session limits for Realtime, which the docs place in account settings, and the maximum session length.", "unchecked: the model pages for `gpt-realtime-2.1` and `gpt-realtime-2.1-mini` (context window, snapshot names), the WebRTC guide, the voice activity detection guide and the full event reference. We stopped at about twenty requests to developers.openai.com, five over the brief's guide of fifteen.", "unchecked: GitHub stars and package download counts. The legal entity was not read. The domain openai.com was registered on 19 January 2007 per RDAP.", "OpenAI shipped GPT-Live (`gpt-live-1`, `/v1/live/sessions`) as generally available on 10 September 2026 and publishes a guide for migrating from Realtime. No shutdown date for the Realtime API was found. GPT-Live is a separate product and is not graded here.", "`gpt-realtime-mini-2025-10-06` was shut down on 23 July 2026 after an announcement on 22 April 2026, three months' notice for a dated snapshot. Whether that snapshot counts as a generally available model under the six-month policy is not clear.", "The lead was right on the interface and the example model. It gave openai.com as the URL, and the documentation and reference are on developers.openai.com." ] }, "negative": 0, "verdict": "A generally available speech-to-speech API with WebRTC, WebSocket and SIP transports, a public OpenAPI file that types its events, and per-token prices. Per-model rate limits appear only in account settings, each turn re-bills the whole conversation, and openai.com refused our reader, so the terms, privacy policy and security pages were not read.", "bestFor": "Teams building their own voice agent on one speech-to-speech model who want browser, server and phone transports from one vendor and can run a backend for secrets and tools.", "strengths": [ "The OpenAPI 3.1 file in `openai/openai-openapi` covers nine Realtime paths and types 13 client events and 48 server events as schemas.", "Browser clients use client secrets that expire after 10 seconds to 2 hours, 10 minutes by default, so the API key stays on a server.", "Remote MCP tools accept an `allowed_tools` list, and `require_approval` defaults to `always` in the OpenAPI file.", "SIP calls connect at `sip.api.openai.com` with accept, reject, refer and hangup endpoints and a signed incoming-call webhook.", "The data controls page lists `/v1/realtime` with no training use, 30-day abuse logs, no stored application state and Zero Data Retention eligibility." ], "weaknesses": [ "openai.com answered 403 to our reader, so the service terms, privacy policy, sub-processor list and security pages were not read.", "Realtime rate limits are shown only in account settings. The public table gives numbers for GPT-6 model families only.", "Every response re-sends the whole conversation, so later turns cost more. Audio is $32 in and $64 out per 1M tokens on `gpt-realtime-2.1`.", "The status feed names Realtime among affected components in five incidents between 24 July and 6 October 2026, with durations not shown in the feed.", "Session settings attached to a client secret can be overridden by the client connection, per the client secret reference." ], "agentNotes": [ "Create a client secret on a server with `POST /v1/realtime/client_secrets` and give browsers only the `ek_` value. Never ship the API key.", "After a response with MCP calls finishes, send another `response.create`. The API does not create the follow-up response itself.", "Set `truncation` with a `retention_ratio` below 1 and `token_limits.post_instructions` to cap input tokens, and keep instructions and tools unchanged to keep the cache.", "On a WebSocket, stop playback on `input_audio_buffer.speech_started` and send `conversation.item.truncate` with `audio_end_ms`. WebRTC and SIP truncate on the server.", "Use `gpt-realtime-2.1` or `gpt-realtime-2.1-mini`. `gpt-realtime` and `gpt-realtime-mini` shut down on 20 January 2027." ], "metrics": { "kind": "remote", "measured": false }, "reviewCount": 0, "avgRating": 0, "history": [ { "basis": "public evidence", "confidence": "low", "grade": "B", "methodology": "0.4", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 63.3 } ], "editorialScores": { "ergonomics": 76, "maintenance": 52, "payments": 20, "reliability": 67, "schema": 84, "security": 61, "transparency": 58 }, "provenanceScore": 84 }, "connect": { "install": "npm install @openai/agents", "http": "curl -X POST https://api.openai.com/v1/realtime/client_secrets \\\n -H \"Authorization: Bearer $OPENAI_API_KEY\" -H \"Content-Type: application/json\" \\\n -d '{\"session\": {\"type\": \"realtime\", \"model\": \"gpt-realtime-2.1\"}}'" }, "letme": { "capability": "https://letme.dev/voice.agent", "tool": "https://letme.dev/openai-realtime" }, "sameCompany": [ "openai-api", "openai-embeddings", "openai-guardrails", "openai-moderation", "openai-image-api", "openai-sora", "openai-speech-to-text", "openai-agents-sdk", "openai-decisions-api", "openai-codex" ], "notable": [ "The Realtime API went generally available on 28 August 2025, and the beta interface (`OpenAI-Beta: realtime=v1`) was removed on 12 May 2026 (https://developers.openai.com/api/docs/changelog.md)", "`gpt-realtime-2.1` and `gpt-realtime-2.1-mini` were released on 6 July 2026 with changes to alphanumeric recognition, noise handling and interruption behaviour (https://developers.openai.com/api/docs/changelog.md)", "`gpt-realtime`, `gpt-4o-realtime`, `gpt-realtime-mini` and `gpt-4o-mini-realtime` shut down on 20 January 2027, notified on 20 July 2026 (https://developers.openai.com/api/docs/deprecations.md)", "GPT-Live (`gpt-live-1`, `/v1/live/sessions`) went generally available on 10 September 2026 as a separate voice API at $0.05 a minute, with a guide for migrating from Realtime (https://developers.openai.com/api/docs/guides/live-migration.md)", "Remote MCP tools are executed by the Realtime API itself, and the client must send a new `response.create` after MCP calls finish (https://developers.openai.com/api/docs/guides/realtime-mcp.md)", "`connector_id` is deprecated for models released after 1 September 2026, with `server_url` or a Secure MCP Tunnel as the replacement (https://developers.openai.com/api/docs/guides/realtime-mcp.md)", "The status feed names Realtime among affected components in five resolved incidents between 24 July and 6 October 2026, while the page shows 100 per cent uptime for the component (https://status.openai.com/)", "openai.com answered 403 to our reader on 9 October 2026, so the service terms, privacy policy and security pages were not read" ], "area": "voice", "details": [ { "label": "Architecture", "value": "Speech-to-speech. The model takes audio in and speaks audio out in one stateful session, with text and image input, optional transcripts and server or semantic voice activity detection" }, { "label": "Endpoints", "value": "`wss://api.openai.com/v1/realtime?model=gpt-realtime-2.1` for WebSocket, `POST /v1/realtime/calls` for WebRTC, `sip:$PROJECT_ID@sip.api.openai.com;transport=tls` for SIP (`sip-eu.api.openai.com` for European data residency), and `POST /v1/realtime/client_secrets` for browser credentials" }, { "label": "Models", "value": "`gpt-realtime-2.1` and `gpt-realtime-2.1-mini` (6 July 2026), `gpt-realtime-2` and `gpt-realtime-1.5`. `gpt-realtime` and `gpt-realtime-mini` shut down on 20 January 2027. `gpt-realtime-translate` and `gpt-realtime-whisper` run translation and transcription sessions billed by the minute" }, { "label": "Tools", "value": "Function tools run by the client or server, and remote MCP tools run by the API with `server_url`, `allowed_tools` and `require_approval` (default `always` in the OpenAPI file). Tools can be set for the session or for one response" }, { "label": "Telephony", "value": "SIP over TLS on port 5061 with SRTP media, a `realtime.call.incoming` webhook, and accept, reject, refer and hangup endpoints under `/v1/realtime/calls/{call_id}`. DTMF events arrive on a sideband connection" }, { "label": "Credentials", "value": "API key as a Bearer header from a server. Client secrets (`ek_`) for browsers, 10 seconds to 2 hours, default 10 minutes" }, { "label": "Billing", "value": "Per token by modality on each response, with the whole conversation re-sent each turn. User audio is 1 token per 100 ms and assistant audio 1 token per 50 ms. Automatic prompt caching, best effort" }, { "label": "Context control", "value": "`truncation` with `retention_ratio` and `token_limits.post_instructions`, `conversation.item.delete`, `conversation.item.truncate` and `max_output_tokens`" }, { "label": "Errors", "value": "An `error` server event with type, code, message, param and the `event_id` of the client event at fault. `mcp_list_tools.failed` and `response.mcp_call.failed` for MCP. REST calls return 429 and 503 with `Retry-After` when present" }, { "label": "Data handling", "value": "Per the data controls page, `/v1/realtime` data is not used for training, abuse monitoring logs are kept up to 30 days, no application state is stored, and Zero Data Retention is available with approval. Tracing is not EU data residency compliant" }, { "label": "SDKs", "value": "Agents SDK for TypeScript, `@openai/agents` and `@openai/agents-realtime` 0.20.0 (8 October 2026), MIT. Docs examples also use the `openai` libraries for JavaScript, Python, Java, Ruby and .NET" }, { "label": "Status", "value": "status.openai.com has a Realtime component under APIs, showing 100 per cent uptime for July to October 2026" } ], "unitPrices": [ { "item": "gpt-realtime-2.1 audio input", "unit": "1m-tokens", "usd": 32, "note": "$0.40 cached. 1 token per 100 ms of user audio, about $0.019 a minute before caching" }, { "item": "gpt-realtime-2.1 audio output", "unit": "1m-tokens", "usd": 64, "note": "1 token per 50 ms of assistant audio, about $0.077 a minute" }, { "item": "gpt-realtime-2.1 text input", "unit": "1m-tokens", "usd": 4, "note": "$0.40 cached" }, { "item": "gpt-realtime-2.1 text output", "unit": "1m-tokens", "usd": 24 }, { "item": "gpt-realtime-2.1-mini audio input", "unit": "1m-tokens", "usd": 10, "note": "$0.30 cached" }, { "item": "gpt-realtime-2.1-mini audio output", "unit": "1m-tokens", "usd": 20 } ], "provenance": { "legalEntity": "OpenAI", "domain": "openai.com", "domainRegistered": "2007-01-19", "endpointOnVendorDomain": true, "terms": "https://openai.com/policies/services-agreement/", "privacy": "https://openai.com/policies/privacy-policy/", "statusPage": "https://status.openai.com", "changelog": "https://developers.openai.com/api/docs/changelog", "securityTxt": "unknown", "checked": "2026-10-09", "notes": [ "The API, WebRTC and SIP endpoints are on api.openai.com and sip.api.openai.com. The docs are on developers.openai.com.", "RDAP for openai.com gives a registration date of 2007-01-19.", "The legal entity was not read. The SDK licence file says Copyright (c) 2025 OpenAI.", "robots.txt answered 200 on developers.openai.com and openai.com, allowing the pages read, and 404 on status.openai.com, which means no rules.", "openai.com answered our researcher with a bot check on 9 October 2026, so the terms and privacy policy were not read on that day. The links are the two documents OpenAI's other listings here carry. openai.com answers our policy reader with HTTP 403 as well, so neither document has been read and both are recorded as unreadable." ], "score": 84, "checks": [ { "check": "Legal entity named", "value": "OpenAI", "points": 20, "max": 20, "state": "ok" }, { "check": "Domain age", "value": "openai.com, registered 2007-01-19 (19 years)", "points": 15, "max": 15, "state": "ok" }, { "check": "Endpoint on the vendor's domain", "value": "api.openai.com", "points": 15, "max": 15, "state": "ok" }, { "check": "Terms of service", "value": "published, but our reader couldn't read it", "points": 7, "max": 10, "state": "part" }, { "check": "Privacy policy", "value": "published, but our reader couldn't read it", "points": 7, "max": 10, "state": "part" }, { "check": "Status page", "value": "status.openai.com", "points": 10, "max": 10, "state": "ok" }, { "check": "Changelog", "value": "published", "points": 10, "max": 10, "state": "ok" }, { "check": "security.txt", "value": "could not be fetched", "points": 0, "max": 10, "state": "no" } ], "policies": [ { "kind": "terms", "url": "https://openai.com/policies/services-agreement/", "state": "unreadable", "reason": "the page answered HTTP 403 to our reader", "readAt": "2026-10-08", "points": 7, "max": 10 }, { "kind": "privacy", "url": "https://openai.com/policies/privacy-policy/", "state": "unreadable", "reason": "the page answered HTTP 403 to our reader", "readAt": "2026-10-08", "points": 7, "max": 10 } ] }, "pageJsonUrl": "https://www.anchorterminal.com/tools/openai-realtime.json", "live": { "slug": "openai-realtime", "probe": { "target": "https://api.openai.com/v1/realtime", "method": "get", "lastAt": "2026-10-10T01:38:01.398025198Z", "lastOk": true, "lastStatus": 405, "lastMs": 147, "authRequired": false, "uptime24h": 100, "uptime30d": 100, "p50ms24h": 153, "p95ms24h": 201, "samples24h": 102, "samples30d": 102, "days": [ { "date": "2026-10-09", "probes": 85, "ok": 85 }, { "date": "2026-10-10", "probes": 17, "ok": 17 } ] }, "vendorStatus": { "page": "https://status.openai.com", "indicator": "minor", "summary": "Partial System Degradation", "checkedAt": "2026-10-10T01:33:58.722210936Z" }, "versions": [ { "registry": "github", "name": "openai/openai-agents-js", "version": "@openai/agents-extensions@0.20.0", "released": "2026-10-08", "seenAt": "2026-10-09T17:10:29.452928495Z" }, { "registry": "npm", "name": "@openai/agents", "version": "0.20.0", "seenAt": "2026-10-09T17:10:27.202909251Z" }, { "registry": "npm", "name": "@openai/agents-realtime", "version": "0.20.0", "seenAt": "2026-10-09T17:10:27.531105633Z" } ], "githubStars": 3903, "npmWeekly": 2125326, "pages": [ { "url": "https://developers.openai.com/api/docs/pricing.md", "kind": "pricing", "status": 200, "checkedAt": "2026-10-09T18:36:10.419663305Z", "changedAt": "0001-01-01T00:00:00Z", "fingerprint": "742c4d3c6b71" } ], "updatedAt": "2026-10-10T01:38:01.398025198Z" } }, "verify": { "accepts": "a page on openai.com or one of its subdomains, or the README of github.com/openai/openai-agents-js", "badgeUrl": "https://www.anchorterminal.com/badges/openai-realtime.svg", "body": { "slug": "openai-realtime", "url": "the page with the badge or the link" }, "docs": "https://www.anchorterminal.com/builders/#verify", "effect": "none, it never changes a grade, rank or review", "endpoint": "https://www.anchorterminal.com/api/v1/verify", "listingUrl": "https://www.anchorterminal.com/tools/openai-realtime", "mcpTool": "verify_listing", "recheck": "weekly; two failed checks in a row and it lapses, a later pass restores it", "snippets": { "html": "\u003ca href=\"https://www.anchorterminal.com/tools/openai-realtime\"\u003e\u003cimg src=\"https://www.anchorterminal.com/badges/openai-realtime.svg\" alt=\"OpenAI Realtime API on Anchor Terminal\" height=\"20\"\u003e\u003c/a\u003e", "markdown": "[![OpenAI Realtime API on Anchor Terminal](https://www.anchorterminal.com/badges/openai-realtime.svg)](https://www.anchorterminal.com/tools/openai-realtime)", "link": "\u003ca href=\"https://www.anchorterminal.com/tools/openai-realtime\"\u003eOpenAI Realtime API on Anchor Terminal\u003c/a\u003e" } } }, "kind": "anchor.page", "links": { "api": "https://www.anchorterminal.com/api/v1/index.json", "html": "https://www.anchorterminal.com/tools/openai-realtime", "json": "https://www.anchorterminal.com/tools/openai-realtime.json", "llms": "https://www.anchorterminal.com/llms.txt", "markdown": "https://www.anchorterminal.com/tools/openai-realtime.md", "slim": "https://www.anchorterminal.com/tools/openai-realtime.min.md" }, "markdown": "## Overview\n\n**Grade B · 63.3/100 · rank #400 of 950 · #7 in Conversational voice agents · not agent-ready · confidence low**\n\n\nMore from OpenAI, listed separately because each is its own product: [OpenAI API](https://www.anchorterminal.com/tools/openai-api.md) (Model APIs \u0026 inference), [OpenAI embeddings](https://www.anchorterminal.com/tools/openai-embeddings.md) (Embeddings \u0026 rerankers), [OpenAI Guardrails](https://www.anchorterminal.com/tools/openai-guardrails.md) (Guardrails \u0026 safety filters), [OpenAI Moderation API](https://www.anchorterminal.com/tools/openai-moderation.md) (Guardrails \u0026 safety filters), [OpenAI Image API](https://www.anchorterminal.com/tools/openai-image-api.md) (Image generation), [OpenAI Sora API](https://www.anchorterminal.com/tools/openai-sora.md) (Video generation), [OpenAI Speech to Text](https://www.anchorterminal.com/tools/openai-speech-to-text.md) (Speech-to-text), [OpenAI Agents SDK](https://www.anchorterminal.com/tools/openai-agents-sdk.md) (Agent frameworks \u0026 SDKs), [OpenAI Decisions API](https://www.anchorterminal.com/tools/openai-decisions-api.md) (Decision models), [OpenAI Codex](https://www.anchorterminal.com/tools/openai-codex.md) (Agent harnesses).\n\n## Assessment\n\nA generally available speech-to-speech API with WebRTC, WebSocket and SIP transports, a public OpenAPI file that types its events, and per-token prices. Per-model rate limits appear only in account settings, each turn re-bills the whole conversation, and openai.com refused our reader, so the terms, privacy policy and security pages were not read.\n\n## Facts\n\n| Field | Value |\n| --- | --- |\n| Vendor | OpenAI (https://openai.com) |\n| Kind | HTTP API |\n| Category | Conversational voice agents (https://www.anchorterminal.com/categories/voice-agents) |\n| Transport | HTTP |\n| Endpoint | `https://api.openai.com/v1/realtime` |\n| Auth | API key · Self-serve API key from platform.openai.com, sent as a Bearer header on server WebSocket connections and call-control requests. For browsers and phones, a server calls `POST /v1/realtime/client_secrets` for a client secret (`ek_`) that lasts 10 seconds to 2 hours, 10 minutes by default, and can start several sessions until it expires. The browser WebSocket example passes the secret as a subprotocol. SIP calls are tied to a project by the project ID in the SIP address. |\n| Pricing | Pay per use (Pay per use) · Per 1M tokens on `gpt-realtime-2.1`. Audio $32.00 in, $0.40 cached and $64.00 out, text $4.00 in and $24.00 out, image $5.00 in. `gpt-realtime-2.1-mini` is $10.00 and $20.00 for audio. Each response re-sends the whole conversation, and input transcription is billed separately. Usage tiers start at $5 of credit purchases. Whether the Free tier covers Realtime models was not established (https://developers.openai.com/api/docs/pricing.md, checked 2026-10-09). |\n| x402 | No · No x402, MPP or L402 in the Realtime guides, the pricing page or the OpenAPI file (checked 2026-10-09). |\n| Licence | Proprietary service. The service terms on openai.com were not read. The Agents SDK for TypeScript and the OpenAPI description are MIT |\n| Packages | npm: `@openai/agents`; npm: `@openai/agents-realtime` |\n| Source | https://github.com/openai/openai-agents-js |\n| Docs | https://developers.openai.com/api/docs/guides/realtime |\n| llms.txt | https://developers.openai.com/api/docs/llms.txt |\n| Last release | 2026-07-28 |\n| Architecture | Speech-to-speech. The model takes audio in and speaks audio out in one stateful session, with text and image input, optional transcripts and server or semantic voice activity detection |\n| Endpoints | `wss://api.openai.com/v1/realtime?model=gpt-realtime-2.1` for WebSocket, `POST /v1/realtime/calls` for WebRTC, `sip:$PROJECT_ID@sip.api.openai.com;transport=tls` for SIP (`sip-eu.api.openai.com` for European data residency), and `POST /v1/realtime/client_secrets` for browser credentials |\n| Models | `gpt-realtime-2.1` and `gpt-realtime-2.1-mini` (6 July 2026), `gpt-realtime-2` and `gpt-realtime-1.5`. `gpt-realtime` and `gpt-realtime-mini` shut down on 20 January 2027. `gpt-realtime-translate` and `gpt-realtime-whisper` run translation and transcription sessions billed by the minute |\n| Tools | Function tools run by the client or server, and remote MCP tools run by the API with `server_url`, `allowed_tools` and `require_approval` (default `always` in the OpenAPI file). Tools can be set for the session or for one response |\n| Telephony | SIP over TLS on port 5061 with SRTP media, a `realtime.call.incoming` webhook, and accept, reject, refer and hangup endpoints under `/v1/realtime/calls/{call_id}`. DTMF events arrive on a sideband connection |\n| Credentials | API key as a Bearer header from a server. Client secrets (`ek_`) for browsers, 10 seconds to 2 hours, default 10 minutes |\n| Billing | Per token by modality on each response, with the whole conversation re-sent each turn. User audio is 1 token per 100 ms and assistant audio 1 token per 50 ms. Automatic prompt caching, best effort |\n| Context control | `truncation` with `retention_ratio` and `token_limits.post_instructions`, `conversation.item.delete`, `conversation.item.truncate` and `max_output_tokens` |\n| Errors | An `error` server event with type, code, message, param and the `event_id` of the client event at fault. `mcp_list_tools.failed` and `response.mcp_call.failed` for MCP. REST calls return 429 and 503 with `Retry-After` when present |\n| Data handling | Per the data controls page, `/v1/realtime` data is not used for training, abuse monitoring logs are kept up to 30 days, no application state is stored, and Zero Data Retention is available with approval. Tracing is not EU data residency compliant |\n| SDKs | Agents SDK for TypeScript, `@openai/agents` and `@openai/agents-realtime` 0.20.0 (8 October 2026), MIT. Docs examples also use the `openai` libraries for JavaScript, Python, Java, Ruby and .NET |\n| Status | status.openai.com has a Realtime component under APIs, showing 100 per cent uptime for July to October 2026 |\n| Capabilities | voice.agent, voice.speech-to-speech, voice.tools, voice.telephony |\n| Tags | hosted, usage-priced, closed-source, speech-to-speech, webrtc, websocket, sip, streaming, typescript, python, openapi, llms-txt, api-key, function-calling, mcp-client, status-page |\n| JSON | https://www.anchorterminal.com/api/v1/tools/openai-realtime.json |\n\n## Score breakdown (methodology v0.4, October 2026 research run)\n\nAssessed 2026-10-09 from public evidence against the published checklist (https://www.anchorterminal.com/benchmark/#checklist). Confidence: low. Performance and Task success pending (no score, not in the total); the total is Σ(score × weight) ÷ 80 over the 7 assessed categories. \"This run\" is each category's share of the 100 points.\n\n| Category | Weight | This run | Score (0–100) | Points |\n| --- | --- | --- | --- | --- |\n| Reliability | 16% | 20 | 67 | 13.4 |\n| Performance | 10% | pending | pending | n/a |\n| Schema \u0026 documentation | 13% | 16.2 | 84 | 13.7 |\n| Agent ergonomics | 13% | 16.2 | 76 | 12.3 |\n| Security \u0026 auth | 14% | 17.5 | 61 | 10.7 |\n| Payments \u0026 pricing | 10% | 12.5 | 20 | 2.5 |\n| Task success | 10% | pending | pending | n/a |\n| Maintenance \u0026 community | 7% | 8.8 | 52 | 4.5 |\n| Transparency \u0026 trust (editorial 58, provenance 84) | 7% | 8.8 | 71 | 6.2 |\n| Negative events | up to −15 | up to −15 | none recorded | 0 |\n| **Total** | | | | **63.3 → B** |\n\n### Why each score\n\n- Reliability 67: Graded as a hosted API. status.openai.com lists Realtime as one of 13 API components with history (20). The page shows 100 per cent uptime for the Realtime component for July to October 2026, and the status feed names Realtime among the affected components of five resolved incidents on 24 July, 25 July (two), 17 September and 6 October. The feed gives only the final update, so durations were not read, and we score them as minor (20 of 30). The rate limits page gives usage tiers and monthly usage limits, but per-model limits for Realtime are shown only in account settings and the public table covers GPT-6 families only (5 of 15). The docs describe `Retry-After` on 429 and 503, exponential backoff with jitter and SDK retries, sessions emit `rate_limits.updated`, and SIP webhooks carry a `webhook-id` for deduplication. No idempotency key for call control was found (12 of 15). No SLA was found in the pages read, and openai.com answered 403 (0). The Realtime API has been generally available since 28 August 2025 (10).\n- Performance: Pending. Latency is measured per call by our probes, which haven't run yet, so this run doesn't score it. Its weight is shared across the assessed categories until the first probe window closes.\n- Schema \u0026 documentation 84: The OpenAPI 3.1 file in `openai/openai-openapi` (updated 8 October 2026) has nine `/realtime` paths and 167 Realtime schemas, with 13 client events and 48 server events typed. It is not an AsyncAPI document, so the WebSocket channel itself is not described (22 of 25). llms.txt and a Markdown twin of every docs page (10). The tools guide has a table of when to use `function`, `mcp` with `server_url` or a connector, and the transport guides say when to choose WebRTC, WebSocket or SIP (16 of 20). Session fields carry enums and ranges, such as `expires_after.seconds` from 10 to 7200, though `model` also accepts any string (13 of 15). Examples in JavaScript, Python, Java, Ruby and C#, an `error` event with type, code, message and param, and a list of common MCP failures. No catalogue of Realtime error codes was found (11 of 15). Dated changelog and dated model snapshots, less 3 because the OpenAPI file still lists `gpt-4o-realtime-preview` models and beta event schemas that the deprecations page says were removed in May 2026, and several guides mix GPT-Live and Realtime text on one page (12 of 15).\n- Agent ergonomics 76: Scored as a model API. Context can be capped with `truncation`, `retention_ratio` and `token_limits.post_instructions`, items can be deleted, prompt caching is automatic, and `response.done` reports token usage by modality (22 of 25). There are no lists to page. Output controls are `output_modalities`, `max_output_tokens`, optional input transcription and `allowed_tools` (14 of 20). The `error` event returns the `event_id` of the client event that caused it, and MCP tool listing and calls have their own failure events (15 of 20). Client events take an `event_id`, webhooks carry a `webhook-id`, and interruptions cancel the response. No session resumption after a dropped connection and no idempotency key for call control were found in the pages read (10 of 20). A session needs only a model, voice activity detection is automatic, and the Agents SDK starts a browser voice agent in about ten lines, with examples in five languages (15).\n- Security \u0026 auth 61: Server connections send the API key as a Bearer header. Browsers use client secrets from `POST /v1/realtime/client_secrets` that last 10 seconds to 2 hours, 10 minutes by default, and can open several sessions until expiry. The browser WebSocket example passes the secret as a subprotocol, not in the query string. The docs index also lists workload identity federation, mutual TLS and an IP allowlist. Key scopes were not read (24 of 30). MCP tools take `allowed_tools` and `require_approval`, which defaults to `always` in the OpenAPI file, and a sideband connection keeps tool execution on the server. Session settings attached to a client secret can be overridden by the client connection (15 of 20). The model hears callers and reads MCP results. The SIP guide says to treat SIP headers as untrusted, and no prompt-injection guidance specific to Realtime was found (6 of 15). The docs index describes Admin APIs with audit log retrieval, sessions can write traces, and usage is reported per response (10 of 15). openai.com answered 403, so security.txt, any bug bounty and certifications are unchecked. We read a private vulnerability reporting policy on the OpenAPI repository and a 5 October changelog entry on HIPAA support (6 of 20).\n- Payments \u0026 pricing 20: No x402, MPP or L402 in the docs or the OpenAPI file (0 of 40). Per-token prices for every Realtime model are published without a login (20). The rate limits page lists a Free tier with a $100 monthly usage limit, but whether it covers Realtime models without a card was not established, and the error guide ties access to prepaid credits (0 of 20). API keys are created in the platform settings after a browser sign-in (0).\n- Task success: Pending. Task success needs the category task suites run through each tool, which haven't run yet, so this run doesn't score it. Its weight is shared across the assessed categories until then. A data provider's data-quality score is published on its listing now and becomes half of this category when it's scored.\n- Maintenance \u0026 community 52: The last dated changelog entry tagged `v1/realtime` that adds a feature is 28 July 2026, 73 days before the check, when `gpt-transcribe` became available for final transcripts of Realtime turns. `gpt-realtime-2.1` shipped on 6 July (20 of 30). Two entries tagged `v1/realtime` fall in the last 90 days, 28 July and 26 August, one short of the three the checklist asks for (0 of 20). Closed service with a changelog updated several times a week, plus a community forum and Discord linked from the docs, which we did not test (10 of 15). Current official SDKs, with `@openai/agents-realtime` 0.20.0 tagged on 8 October 2026 and 12 version tags since 29 July (15). The SDK is MIT but still 0.x, and its CI results were not read (7 of 10).\n- Transparency \u0026 trust 71: Closed service. The SDKs and the OpenAPI file are MIT. The service terms were not read because openai.com answered 403 (10 of 30). The data controls page says API data is not used for training unless the customer opts in, abuse monitoring logs are kept up to 30 days, `/v1/realtime` stores no application state and is eligible for Zero Data Retention with approval. The privacy policy and any DPA were not read, so agreement between them is unchecked (18 of 30). The deprecations page states at least six months' notice for generally available models and lists shutdown dates with replacements (20 of 20). Data residency exists for the United States and Europe, with a `sip-eu.api.openai.com` endpoint, and tracing on `/v1/realtime` is not EU residency compliant. The sub-processor list is linked but was not read (10 of 20).\n\nFix list for a coding agent, everything this grade says the listing lacks, the biggest gain first (20 items): https://www.anchorterminal.com/fixes/openai-realtime.md (JSON https://www.anchorterminal.com/fixes/openai-realtime.json)\n\n### What we couldn't check\n\n- unchecked: the service terms, privacy policy, DPA, sub-processor list, security and trust pages and security.txt on openai.com. The host answered 403 to one request on 9 October 2026 and we did not retry. `provenance.terms` and `provenance.privacy` are left out for that reason.\n- unchecked: whether an SLA is published for any tier. None was found in the developer docs. The rate limits page links Scale Tier and Reserved Tier pages on openai.com, which were not read.\n- unchecked: whether the Free usage tier gives access to Realtime models without a card or prepaid credits.\n- unchecked: duration and severity of the five incidents that name Realtime. The status feed shows only the final update of each.\n- unchecked: per-model rate limits and concurrent session limits for Realtime, which the docs place in account settings, and the maximum session length.\n- unchecked: the model pages for `gpt-realtime-2.1` and `gpt-realtime-2.1-mini` (context window, snapshot names), the WebRTC guide, the voice activity detection guide and the full event reference. We stopped at about twenty requests to developers.openai.com, five over the brief's guide of fifteen.\n- unchecked: GitHub stars and package download counts. The legal entity was not read. The domain openai.com was registered on 19 January 2007 per RDAP.\n- OpenAI shipped GPT-Live (`gpt-live-1`, `/v1/live/sessions`) as generally available on 10 September 2026 and publishes a guide for migrating from Realtime. No shutdown date for the Realtime API was found. GPT-Live is a separate product and is not graded here.\n- `gpt-realtime-mini-2025-10-06` was shut down on 23 July 2026 after an announcement on 22 April 2026, three months' notice for a dated snapshot. Whether that snapshot counts as a generally available model under the six-month policy is not clear.\n- The lead was right on the interface and the example model. It gave openai.com as the URL, and the documentation and reference are on developers.openai.com.\n\n### Sources\n\n- Getting started with the Realtime API (Markdown twin): \u003chttps://developers.openai.com/api/docs/guides/realtime.md\u003e (seen 2026-10-09)\n- Realtime with tools, function and MCP tools, approvals: \u003chttps://developers.openai.com/api/docs/guides/realtime-mcp.md\u003e (seen 2026-10-09)\n- Realtime conversations, events, errors, interruption: \u003chttps://developers.openai.com/api/docs/guides/realtime-conversations.md\u003e (seen 2026-10-09)\n- WebSockets guide, authentication for server and browser: \u003chttps://developers.openai.com/api/docs/guides/voice-websockets.md\u003e (seen 2026-10-09)\n- Telephony and SIP guide: \u003chttps://developers.openai.com/api/docs/guides/voice-sip.md\u003e (seen 2026-10-09)\n- Server-side controls, sideband connections: \u003chttps://developers.openai.com/api/docs/guides/voice-server-controls.md\u003e (seen 2026-10-09)\n- Cost guide, Realtime token accounting, caching and truncation: \u003chttps://developers.openai.com/api/docs/guides/voice-latency-cost.md\u003e (seen 2026-10-09)\n- Migrate to GPT-Live guide: \u003chttps://developers.openai.com/api/docs/guides/live-migration.md\u003e (seen 2026-10-09)\n- Pricing: \u003chttps://developers.openai.com/api/docs/pricing.md\u003e (seen 2026-10-09)\n- Rate limits and usage tiers: \u003chttps://developers.openai.com/api/docs/guides/rate-limits.md\u003e (seen 2026-10-09)\n- Error codes: \u003chttps://developers.openai.com/api/docs/guides/error-codes.md\u003e (seen 2026-10-09)\n- Deprecations and notice periods: \u003chttps://developers.openai.com/api/docs/deprecations.md\u003e (seen 2026-10-09)\n- API changelog: \u003chttps://developers.openai.com/api/docs/changelog.md\u003e (seen 2026-10-09)\n- Data controls, retention and residency: \u003chttps://developers.openai.com/api/docs/guides/your-data.md\u003e (seen 2026-10-09)\n- Create client secret reference: \u003chttps://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create.md\u003e (seen 2026-10-09)\n- Docs index (llms.txt) for the API guides: \u003chttps://developers.openai.com/api/docs/llms.txt\u003e (seen 2026-10-09)\n- OpenAPI 3.1 description, read as a file from the repository: \u003chttps://github.com/openai/openai-openapi\u003e (seen 2026-10-09)\n- Agents SDK for TypeScript, tags and `@openai/agents-realtime` manifest, read from a clone: \u003chttps://github.com/openai/openai-agents-js\u003e (seen 2026-10-09)\n- Status page, components and uptime: \u003chttps://status.openai.com/\u003e (seen 2026-10-09)\n- Status feed, incidents since July 2026: \u003chttps://status.openai.com/feed.atom\u003e (seen 2026-10-09)\n- RDAP record for openai.com: \u003chttps://rdap.verisign.com/com/v1/domain/openai.com\u003e (seen 2026-10-09)\n- robots.txt for developers.openai.com (200, allows all), openai.com (200, allows all but one path) and status.openai.com (404, so no rules): \u003chttps://developers.openai.com/robots.txt\u003e (seen 2026-10-09)\n\n## Who's behind it (provenance 84/100, checked 2026-10-09)\n\n| Check | Finding | Points |\n| --- | --- | --- |\n| Legal entity named | OpenAI | 20/20 |\n| Domain age | openai.com, registered 2007-01-19 (19 years) | 15/15 |\n| Endpoint on the vendor's domain | api.openai.com | 15/15 |\n| Terms of service | published, but our reader couldn't read it | 7/10 |\n| Privacy policy | published, but our reader couldn't read it | 7/10 |\n| Status page | status.openai.com | 10/10 |\n| Changelog | published | 10/10 |\n| security.txt | could not be fetched | 0/10 |\n\nThe API, WebRTC and SIP endpoints are on api.openai.com and sip.api.openai.com. The docs are on developers.openai.com.\n\nRDAP for openai.com gives a registration date of 2007-01-19.\n\nThe legal entity was not read. The SDK licence file says Copyright (c) 2025 OpenAI.\n\nrobots.txt answered 200 on developers.openai.com and openai.com, allowing the pages read, and 404 on status.openai.com, which means no rules.\n\nopenai.com answered our researcher with a bot check on 9 October 2026, so the terms and privacy policy were not read on that day. The links are the two documents OpenAI's other listings here carry. openai.com answers our policy reader with HTTP 403 as well, so neither document has been read and both are recorded as unreadable.\n\n### Terms and privacy, as read\n\nA reading by a fixed set of rules, each answered with the vendor's own sentence. Not legal advice.\n\n**Terms of service** (https://openai.com/policies/services-agreement/), read 2026-10-08. Our reader couldn't read it (the page answered HTTP 403 to our reader).\n\n\n**Privacy policy** (https://openai.com/policies/privacy-policy/), read 2026-10-08. Our reader couldn't read it (the page answered HTTP 403 to our reader).\n\n\n## Live (updated 2026-10-10 01:38 UTC)\n\n- Right now: up, HTTP 405, 147 ms, checked 2026-10-10 01:38 UTC (get on `https://api.openai.com/v1/realtime`)\n- Uptime 24h 100.0% (102 probes) · 30 days 100.0% (102 probes) · p50 153 ms · p95 201 ms\n- Vendor status page: minor, Partial System Degradation\n- github `openai/openai-agents-js` @openai/agents-extensions@0.20.0, released 2026-10-08\n- npm `@openai/agents` 0.20.0\n- npm `@openai/agents-realtime` 0.20.0\n- Watching pricing \u003chttps://developers.openai.com/api/docs/pricing.md\u003e\n- Always current: https://www.anchorterminal.com/api/v1/live/openai-realtime.json\n\n## Probe metrics\n\nNot measured yet. Our benchmark probes haven't run, so there's no availability, latency or error rate from a run and Performance is pending. Live uptime, where we poll the endpoint, is under Live and doesn't change the score.\n\n## Prices\n\n| Item | Price | Unit | Note |\n| --- | --- | --- | --- |\n| gpt-realtime-2.1 audio input | $32 | per 1M tokens | $0.40 cached. 1 token per 100 ms of user audio, about $0.019 a minute before caching |\n| gpt-realtime-2.1 audio output | $64 | per 1M tokens | 1 token per 50 ms of assistant audio, about $0.077 a minute |\n| gpt-realtime-2.1 text input | $4 | per 1M tokens | $0.40 cached |\n| gpt-realtime-2.1 text output | $24 | per 1M tokens | |\n| gpt-realtime-2.1-mini audio input | $10 | per 1M tokens | $0.30 cached |\n| gpt-realtime-2.1-mini audio output | $20 | per 1M tokens | |\n\nAcross all listings: https://www.anchorterminal.com/prices/index.md\n\n## Strengths\n\n- The OpenAPI 3.1 file in `openai/openai-openapi` covers nine Realtime paths and types 13 client events and 48 server events as schemas.\n- Browser clients use client secrets that expire after 10 seconds to 2 hours, 10 minutes by default, so the API key stays on a server.\n- Remote MCP tools accept an `allowed_tools` list, and `require_approval` defaults to `always` in the OpenAPI file.\n- SIP calls connect at `sip.api.openai.com` with accept, reject, refer and hangup endpoints and a signed incoming-call webhook.\n- The data controls page lists `/v1/realtime` with no training use, 30-day abuse logs, no stored application state and Zero Data Retention eligibility.\n\n## Weaknesses\n\n- openai.com answered 403 to our reader, so the service terms, privacy policy, sub-processor list and security pages were not read.\n- Realtime rate limits are shown only in account settings. The public table gives numbers for GPT-6 model families only.\n- Every response re-sends the whole conversation, so later turns cost more. Audio is $32 in and $64 out per 1M tokens on `gpt-realtime-2.1`.\n- The status feed names Realtime among affected components in five incidents between 24 July and 6 October 2026, with durations not shown in the feed.\n- Session settings attached to a client secret can be overridden by the client connection, per the client secret reference.\n\n## Before you call it (notes for agents)\n\n1. Create a client secret on a server with `POST /v1/realtime/client_secrets` and give browsers only the `ek_` value. Never ship the API key.\n2. After a response with MCP calls finishes, send another `response.create`. The API does not create the follow-up response itself.\n3. Set `truncation` with a `retention_ratio` below 1 and `token_limits.post_instructions` to cap input tokens, and keep instructions and tools unchanged to keep the cache.\n4. On a WebSocket, stop playback on `input_audio_buffer.speech_started` and send `conversation.item.truncate` with `audio_end_ms`. WebRTC and SIP truncate on the server.\n5. Use `gpt-realtime-2.1` or `gpt-realtime-2.1-mini`. `gpt-realtime` and `gpt-realtime-mini` shut down on 20 January 2027.\n\n## Connect\n\nInstall:\n\n```bash\nnpm install @openai/agents\n```\n\nFirst request:\n\n```bash\ncurl -X POST https://api.openai.com/v1/realtime/client_secrets \\\n -H \"Authorization: Bearer $OPENAI_API_KEY\" -H \"Content-Type: application/json\" \\\n -d '{\"session\": {\"type\": \"realtime\", \"model\": \"gpt-realtime-2.1\"}}'\n```\n\nThrough letme (picks today, calling later): https://letme.dev/openai-realtime. letme answers with the pick and how to call it direct; calling through letme (one key, the vendor's own price) comes later. How it works: https://www.anchorterminal.com/letme/index.md\n\n## Similar tools\n\nRanked by shared capabilities, then score. Same-category tools with no shared capability key are listed last.\n\n| Tool | Grade | Score | Rank | Shared capabilities | x402 | Markdown |\n| --- | --- | --- | --- | --- | --- | --- |\n| Retell AI API + MCP | B | 69.1 | 196 | voice.agent, voice.speech-to-speech, voice.tools, voice.telephony | no | https://www.anchorterminal.com/tools/retell-ai.md |\n| Vapi API + MCP | B | 63.5 | 390 | voice.agent, voice.speech-to-speech, voice.tools, voice.telephony | no | https://www.anchorterminal.com/tools/vapi.md |\n| Ultravox Realtime API | C | 58.5 | 576 | voice.agent, voice.speech-to-speech, voice.tools, voice.telephony | no | https://www.anchorterminal.com/tools/ultravox.md |\n| Hume EVI (Empathic Voice Interface) | C | 56.7 | 626 | voice.agent, voice.speech-to-speech, voice.tools, voice.telephony | no | https://www.anchorterminal.com/tools/hume-evi.md |\n| Bolna API + MCP | D | 52.6 | 729 | voice.agent, voice.speech-to-speech, voice.tools, voice.telephony | no | https://www.anchorterminal.com/tools/bolna.md |\n| ElevenLabs Agents API + MCP | BB | 71.3 | 130 | voice.agent, voice.tools, voice.telephony | no | https://www.anchorterminal.com/tools/elevenlabs-agents.md |\n\n## Panel reviews (0)\n\nReviewed by the Anchor panel (https://www.anchorterminal.com/reviewers/index.md): .\n\nDesk reviews, written from public documentation, pricing, terms, source and status history on 1 October 2026. No calls made. For a desk review, the outcome says whether the reviewer's questions could be answered from public material: success, partial or failure. How reviews work: https://www.anchorterminal.com/reviews/how-it-works.md\n\n## Notable\n\n- The Realtime API went generally available on 28 August 2025, and the beta interface (`OpenAI-Beta: realtime=v1`) was removed on 12 May 2026 (source: \u003chttps://developers.openai.com/api/docs/changelog.md\u003e)\n- `gpt-realtime-2.1` and `gpt-realtime-2.1-mini` were released on 6 July 2026 with changes to alphanumeric recognition, noise handling and interruption behaviour (source: \u003chttps://developers.openai.com/api/docs/changelog.md\u003e)\n- `gpt-realtime`, `gpt-4o-realtime`, `gpt-realtime-mini` and `gpt-4o-mini-realtime` shut down on 20 January 2027, notified on 20 July 2026 (source: \u003chttps://developers.openai.com/api/docs/deprecations.md\u003e)\n- GPT-Live (`gpt-live-1`, `/v1/live/sessions`) went generally available on 10 September 2026 as a separate voice API at $0.05 a minute, with a guide for migrating from Realtime (source: \u003chttps://developers.openai.com/api/docs/guides/live-migration.md\u003e)\n- Remote MCP tools are executed by the Realtime API itself, and the client must send a new `response.create` after MCP calls finish (source: \u003chttps://developers.openai.com/api/docs/guides/realtime-mcp.md\u003e)\n- `connector_id` is deprecated for models released after 1 September 2026, with `server_url` or a Secure MCP Tunnel as the replacement (source: \u003chttps://developers.openai.com/api/docs/guides/realtime-mcp.md\u003e)\n- The status feed names Realtime among affected components in five resolved incidents between 24 July and 6 October 2026, while the page shows 100 per cent uptime for the component (source: \u003chttps://status.openai.com/\u003e)\n- openai.com answered 403 to our reader on 9 October 2026, so the service terms, privacy policy and security pages were not read\n\n- #7 of 14 in Best conversational voice-agent APIs: https://www.anchorterminal.com/best/voice-agents/index.md\n- All 91 voice agents comparisons: https://www.anchorterminal.com/compare/voice-agents/index.md\n\n## Compare\n\n- [AssemblyAI Voice Agent API vs OpenAI Realtime API](https://www.anchorterminal.com/compare/assemblyai-voice-agent-vs-openai-realtime.md): B 64.4 vs B 63.3\n- [Bland AI API + MCP vs OpenAI Realtime API](https://www.anchorterminal.com/compare/bland-ai-vs-openai-realtime.md): B 63.8 vs B 63.3\n- [Bolna API + MCP vs OpenAI Realtime API](https://www.anchorterminal.com/compare/bolna-vs-openai-realtime.md): D 52.6 vs B 63.3\n- [Deepgram Voice Agent API vs OpenAI Realtime API](https://www.anchorterminal.com/compare/deepgram-voice-agent-vs-openai-realtime.md): B 68.1 vs B 63.3\n- [ElevenLabs Agents API + MCP vs OpenAI Realtime API](https://www.anchorterminal.com/compare/elevenlabs-agents-vs-openai-realtime.md): BB 71.3 vs B 63.3\n- [Gemini Live API vs OpenAI Realtime API](https://www.anchorterminal.com/compare/gemini-live-vs-openai-realtime.md): C 57.1 vs B 63.3\n- [Hume EVI (Empathic Voice Interface) vs OpenAI Realtime API](https://www.anchorterminal.com/compare/hume-evi-vs-openai-realtime.md): C 56.7 vs B 63.3\n- [OpenAI Realtime API vs Retell AI API + MCP](https://www.anchorterminal.com/compare/openai-realtime-vs-retell-ai.md): B 63.3 vs B 69.1\n- [OpenAI Realtime API vs Synthflow API + MCP](https://www.anchorterminal.com/compare/openai-realtime-vs-synthflow.md): B 63.3 vs D 51.1\n- [OpenAI Realtime API vs Ultravox Realtime API](https://www.anchorterminal.com/compare/openai-realtime-vs-ultravox.md): B 63.3 vs C 58.5\n- [OpenAI Realtime API vs Vapi API + MCP](https://www.anchorterminal.com/compare/openai-realtime-vs-vapi.md): B 63.3 vs B 63.5\n- [OpenAI Realtime API vs Vocily AI](https://www.anchorterminal.com/compare/openai-realtime-vs-vocily.md): B 63.3 vs D 53.8\n- [OpenAI Realtime API vs Vogent API](https://www.anchorterminal.com/compare/openai-realtime-vs-vogent.md): B 63.3 vs D 47.1\n\n## Verify this listing\n\nFor the vendor. The badge or a plain link to this page verifies the listing, from a page on openai.com or one of its subdomains, or the README of github.com/openai/openai-agents-js. It shows the listing is the vendor's and that the vendor knows it's here, and it never changes a grade, rank or review. The vendor sends the page's address to `POST https://www.anchorterminal.com/api/v1/verify` as `{\"slug\": \"openai-realtime\", \"url\": \"…\"}`, or calls the `verify_listing` tool at https://www.anchorterminal.com/mcp. We fetch the page once, then again every week; two failed checks in a row and the verification lapses, and a later pass restores it. What we check: https://www.anchorterminal.com/builders/index.md#verify\n\nHTML badge:\n\n```html\n\u003ca href=\"https://www.anchorterminal.com/tools/openai-realtime\"\u003e\u003cimg src=\"https://www.anchorterminal.com/badges/openai-realtime.svg\" alt=\"OpenAI Realtime API on Anchor Terminal\" height=\"20\"\u003e\u003c/a\u003e\n```\n\nMarkdown badge, for a README:\n\n```markdown\n[![OpenAI Realtime API on Anchor Terminal](https://www.anchorterminal.com/badges/openai-realtime.svg)](https://www.anchorterminal.com/tools/openai-realtime)\n```\n\nPlain link:\n\n```html\n\u003ca href=\"https://www.anchorterminal.com/tools/openai-realtime\"\u003eOpenAI Realtime API on Anchor Terminal\u003c/a\u003e\n```\n\n## Share this listing\n\nFor the vendor. Sharing assets for social media, two PNGs of 1200 × 630 that say OpenAI Realtime API is listed on Anchor Terminal, with the vendor's logo and this page's address and no grade or score.\n\n- Dark: https://www.anchorterminal.com/assets/share/openai-realtime-dark.png\n- Light: https://www.anchorterminal.com/assets/share/openai-realtime-light.png\n", "meta": { "attribution": "Anchor Terminal (https://www.anchorterminal.com)", "docs": "https://www.anchorterminal.com/docs/", "generatedAt": "2026-10-10", "license": "CC-BY-4.0", "method": "https://www.anchorterminal.com/benchmark/", "methodology": "0.4", "openapi": "https://www.anchorterminal.com/openapi.json", "preview": false, "run": "2026-10-01", "runLabel": "October 2026 research run" }, "page": { "breadcrumbs": [ { "name": "Home", "url": "https://www.anchorterminal.com/" }, { "name": "Terminal", "url": "https://www.anchorterminal.com/tools/" }, { "name": "Conversational voice agents", "url": "https://www.anchorterminal.com/categories/voice-agents" }, { "name": "OpenAI Realtime API", "url": "" } ], "description": "OpenAI's Realtime API runs spoken conversations with speech-to-speech models such as gpt-realtime-2.1. Clients connect over WebRTC, WebSocket or SIP, and sessions support interruptions, function calling and remote MCP tools.", "facts": [ "rank #400 of 950", "API key auth", "0 desk reviews" ], "h1": "OpenAI Realtime API", "image": "https://www.anchorterminal.com/assets/og/tools-openai-realtime.png", "path": "/tools/openai-realtime", "published": "2026-10-01", "section": "tools", "title": "OpenAI Realtime review (2026): pricing, alternatives and grade B", "toc": null, "updated": "2026-10-10", "url": "https://www.anchorterminal.com/tools/openai-realtime" }, "tokens": { "markdown": 8300, "slim": 1980 }, "version": 1 }