{ "data": { "a": { "slug": "amazon-nova-sonic", "name": "Amazon Nova 2 Sonic", "vendor": "Amazon Web Services", "vendorUrl": "https://aws.amazon.com/ai/generative-ai/nova/speech/", "kind": "http-api", "category": "voice-agents", "summary": "Amazon Nova 2 Sonic is AWS's speech-to-speech model for voice agents on Amazon Bedrock. Clients stream audio and text events both ways over the `InvokeModelWithBidirectionalStream` API, with turn-taking, interruptions, asynchronous tool calls and multilingual voices.", "url": "https://www.anchorterminal.com/tools/amazon-nova-sonic", "markdownUrl": "https://www.anchorterminal.com/tools/amazon-nova-sonic.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/amazon-nova-sonic.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/amazon-nova-sonic.json", "dns": "amazon-nova-sonic.grade.anchorterminal.com", "repo": "https://github.com/aws-samples/amazon-nova-samples", "transports": [ "http" ], "packages": [ { "registry": "pypi", "name": "aws-sdk-bedrock-runtime" }, { "registry": "npm", "name": "@aws-sdk/client-bedrock-runtime" } ], "auth": "api-key", "authNotes": "AWS Signature Version 4 with IAM credentials or a role, scoped per model ARN by IAM policy. The Sonic quickstart signs with environment credentials. Whether Bedrock API keys work on the bidirectional stream wasn't confirmed. Access is self-serve once a person has an AWS account.", "pricing": "usage", "pricingNotes": "Billed per speech and text token on Bedrock on demand. The Nova and Bedrock pricing pages draw their price tables by script, so the figures weren't read on 6 November 2026. No model-specific free tier; new AWS accounts get up to $200 in Free Tier credits.", "priceSummary": "Pay per use", "where": "local", "x402": { "level": "no", "evidence": "No x402, MPP or L402 in the Nova or Bedrock docs (checked 2026-11-06). Billing runs through the AWS account.", "endpoints": [] }, "toolCount": 0, "popularity": { "githubStars": null, "npmWeekly": null, "pypiWeekly": null, "asOf": "2026-11-06" }, "docsUrl": "https://docs.aws.amazon.com/nova/latest/nova2-userguide/using-conversational-speech.html", "llmsTxt": "https://docs.aws.amazon.com/nova/latest/nova2-userguide/llms.txt", "capabilities": [ "voice.agent", "voice.speech-to-speech", "voice.tools", "voice.telephony" ], "tags": [ "official", "hosted", "usage-priced", "closed-source", "python", "typescript", "llms-txt", "streaming", "enterprise" ], "lastRelease": "2026-05-28", "graded": true, "anchor": { "graded": true, "score": 67.6, "grade": "B", "agentReady": false, "rank": 314, "ranked": true, "rankOf": 1486, "categoryRank": 6, "methodology": "0.7", "run": "2026-10-01", "scores": { "ergonomics": 64, "maintenance": 50, "payments": 10, "reliability": 95, "schema": 75, "security": 77, "transparency": 79 }, "pending": [ "performance", "tasks" ], "assessment": { "confidence": "low", "date": "2026-11-06" }, "negative": 0, "verdict": "Nova 2 Sonic runs speech-to-speech conversations with asynchronous tools on Bedrock, with IAM scoping, no default storage of inputs and a 99.9 per cent SLA. Sessions cut off at 8 minutes, the Python SDK is pre-alpha, and the event payloads have no schema file. Prices are in a script-drawn table we couldn't read.", "bestFor": "Voice agents on AWS, call centres on Amazon Connect, teams that want IAM and in-account data handling.", "strengths": [ "No Bedrock event on the AWS Health history in the 90 days from 8 August 2026", "IAM scoping per model ARN with SigV4 and temporary credentials", "Bedrock stores no inputs or outputs by default, and Sonic isn't on the list of models kept for abuse review", "Asynchronous tool calls let the model keep speaking while a tool runs", "Telephony guides for Twilio, Vonage and Amazon Connect" ], "weaknesses": [ "Connections end after 8 minutes, so long calls need the session-continuation pattern", "The Python SDK for bidirectional streaming is pre-alpha and needs Python 3.12", "Sonic's event payloads are documented as examples, with no schema file", "20 concurrent sessions per Region, not adjustable through Service Quotas", "The March and May 2026 refreshes changed the model in place under the same ID" ], "agentNotes": [ "Plan a reconnect before 8 minutes and replay chat history (up to 200 KB) at the start of the new session", "Send session, prompt and content start events in order before any audio", "Make tool calls safe to repeat, since a reconnect can replay them", "Back off on `ThrottlingException` and stay within 20 concurrent sessions per Region", "Use the polyglot voices TIFFANY or MATTHEW when callers may switch language" ], "metrics": { "kind": "remote", "measured": false }, "reviewCount": 0, "avgRating": 0, "history": [ { "basis": "public evidence", "confidence": "low", "grade": "B", "methodology": "0.7", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 67.6 } ], "editorialScores": { "ergonomics": 64, "maintenance": 50, "payments": 10, "reliability": 95, "schema": 75, "security": 77, "transparency": 64 }, "provenanceScore": 94, "checked": { "categories": { "ergonomics": "2026-11-06", "maintenance": "2026-11-06", "payments": "2026-11-06", "reliability": "2026-11-06", "schema": "2026-11-06", "security": "2026-11-06", "transparency": "2026-11-06" }, "last": "2026-11-06", "oldest": "2026-11-06", "stale": false, "staleAfterDays": 90 } }, "connect": { "install": "pip install aws-sdk-bedrock-runtime" }, "letme": { "capability": "https://letme.dev/voice.agent", "tool": "https://letme.dev/amazon-nova-sonic" }, "sameCompany": [ "amazon-bedrock", "amazon-nova-embeddings", "amazon-bedrock-guardrails", "amazon-transcribe", "amazon-polly", "agentcore-memory", "agentcore-identity", "aws-secrets-manager", "agentcore-web-search", "aws-mcp-servers", "amazon-ses", "amazon-location", "amazon-translate", "amazon-ads-api", "amazon-creators-api" ], "area": "voice", "provenance": { "legalEntity": "Amazon Web Services, Inc.", "domain": "amazon.com", "domainRegistered": "1994-11-01", "domainNote": "The endpoint is on amazonaws.com, an AWS domain registered on 2005-08-18. The security.txt on aws.amazon.com expires on 2027-10-16.", "endpointOnVendorDomain": true, "terms": "https://aws.amazon.com/service-terms/", "privacy": "https://aws.amazon.com/privacy/", "statusPage": "https://health.aws.amazon.com/health/status", "changelog": "https://docs.aws.amazon.com/nova/latest/nova2-userguide/release-notes.html", "securityTxt": "valid", "checked": "2026-11-06", "notes": [ "The AWS Service Terms were last updated on 30 October 2026. Section 50.12 covers Amazon Bedrock, and Bedrock is not among the services section 50.3 lists for use of content to improve AWS services.", "RDAP gives 1994-11-01 for amazon.com and 2005-08-18 for amazonaws.com.", "aws.amazon.com/.well-known/security.txt has Contact, Policy and Encryption fields. Its Expires value is 2027-10-16T16:25:03Z, re-read on 2026-11-01.", "docs.aws.amazon.com/robots.txt allows the Nova and Bedrock pages read. The Nova and Bedrock pricing pages draw their tables by script, so Nova 2 Sonic's prices weren't read." ], "score": 94 }, "pageJsonUrl": "https://www.anchorterminal.com/tools/amazon-nova-sonic.json", "live": { "slug": "amazon-nova-sonic", "versions": [ { "registry": "npm", "name": "@aws-sdk/client-bedrock-runtime", "version": "3.1164.0", "seenAt": "2026-11-06T23:05:35.366071006Z" }, { "registry": "pypi", "name": "aws-sdk-bedrock-runtime", "version": "0.12.0", "released": "2026-10-08", "seenAt": "2026-11-06T23:05:33.518358123Z" } ], "githubStars": 456, "npmWeekly": 21825916, "pypiWeekly": 112826, "securityTxt": { "url": "https://amazon.com/.well-known/security.txt", "state": "valid", "checkedAt": "2026-11-07T16:00:18.354437696Z" }, "llmsTxt": { "url": "https://docs.aws.amazon.com/nova/latest/nova2-userguide/llms.txt", "ok": true, "status": 200, "checkedAt": "2026-11-07T15:31:25.828159373Z" }, "pages": [ { "url": "https://docs.aws.amazon.com/nova/latest/nova2-userguide/release-notes.html", "kind": "changelog", "status": 200, "checkedAt": "2026-11-07T14:05:20.262718007Z", "changedAt": "0001-01-01T00:00:00Z", "fingerprint": "48cb3eb6c070" } ], "updatedAt": "2026-11-07T16:00:18.354437696Z" } }, "answer": "OpenAI Realtime API scores 71.7 (BB) on agent readiness against Amazon Nova 2 Sonic's 67.6 (B), and leads in 6 of 7 scored categories. Amazon Nova 2 Sonic leads on reliability.", "b": { "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", "dns": "openai-realtime.grade.anchorterminal.com", "repo": "https://github.com/openai/openai-agents-js", "license": "Proprietary service under OpenAI's services agreement, per the OpenAI API listing's check (openai.com answered 403 to our reader on 2 November 2026). 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, and the model page's rate limits start at Build, with no Free row (https://developers.openai.com/api/docs/pricing.md, checked 2026-11-02).", "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": "2026-10-20", "graded": true, "anchor": { "graded": true, "score": 71.7, "grade": "BB", "agentReady": true, "rank": 148, "ranked": true, "rankOf": 1486, "categoryRank": 4, "methodology": "0.7", "run": "2026-10-01", "scores": { "ergonomics": 76, "maintenance": 82, "payments": 30, "reliability": 62, "schema": 84, "security": 83, "transparency": 90 }, "pending": [ "performance", "tasks" ], "assessment": { "confidence": "medium", "date": "2026-11-02" }, "negative": 0, "verdict": "A generally available speech-to-speech API with WebRTC, WebSocket and SIP transports, a public OpenAPI file that types its events, short-lived browser secrets and per-token prices. Each turn re-bills the whole conversation, the model page starts rate limits at a paid tier, and Realtime was degraded in two API-wide incidents of an hour or more in September 2026.", "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 11 client events and 46 server events in its event unions.", "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": [ "Realtime was degraded in two API-wide incidents, for an hour on 17 September and for 5 hours 21 minutes on 29 September 2026.", "The `gpt-realtime-2.1` rate-limit table starts at Build, which needs $5 of credit, so Free tier access to Realtime isn't established.", "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`.", "Session settings attached to a client secret can be overridden by the client connection, per the client secret reference.", "Realtime sits under one Model capabilities permission, so a restricted key can't be limited to Realtime alone." ], "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, and `whisper-1` and `gpt-4o-transcribe` on 26 February 2027." ], "metrics": { "kind": "remote", "measured": false }, "reviewCount": 4, "avgRating": 3.3, "history": [ { "basis": "public evidence", "confidence": "medium", "grade": "BB", "methodology": "0.7", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 71.7 } ], "editorialScores": { "ergonomics": 76, "maintenance": 82, "payments": 30, "reliability": 62, "schema": 84, "security": 83, "transparency": 85 }, "provenanceScore": 95, "checked": { "categories": { "ergonomics": "2026-11-02", "maintenance": "2026-11-02", "payments": "2026-11-02", "reliability": "2026-11-02", "schema": "2026-11-02", "security": "2026-11-02", "transparency": "2026-11-02" }, "last": "2026-11-02", "lastMoved": "2026-11-02", "oldest": "2026-11-02", "stale": false, "staleAfterDays": 90 } }, "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", "gpt-live", "openai-agents-sdk", "openai-decisions-api", "openai-agents-api", "openai-codex", "gpt-oss-120b", "gpt-oss-20b" ], "area": "voice", "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 OpCo, LLC", "domain": "openai.com", "domainRegistered": "2007-01-19", "domainNote": "openai.com was registered in 2007, before OpenAI existed.", "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": "valid", "checked": "2026-11-02", "notes": [ "The API, WebRTC and SIP endpoints are on api.openai.com and sip.api.openai.com. The docs are on developers.openai.com.", "Same entity, terms and privacy policy as the OpenAI API listing. The legal entity and the two policy URLs stand per that listing's check of 26 September. openai.com answered the services agreement with HTTP 403 on 2 November 2026.", "RDAP gives openai.com a registration date of 2007-01-19.", "openai.com/.well-known/security.txt, read on 2 November 2026, is PGP-signed with Bugcrowd and email contacts and a disclosure policy, and has no Expires field.", "robots.txt answered 200 on developers.openai.com, openai.com and trust.openai.com, allowing the pages read, and 404 on status.openai.com, which means no rules.", "status.openai.com lists a Realtime component under the API group." ], "score": 95 }, "requirements": { "setupChecked": true, "complianceChecked": true, "card": "paid-only", "freeTier": true, "minutes": 5, "credential": "api-key", "hosting": [ "US", "EU", "UK", "CA", "APAC" ], "euResidency": "enterprise", "trainsOnData": "no", "zeroRetention": "enterprise", "certifications": [ "SOC 2 Type II", "ISO 27001", "ISO 27701", "ISO 42001", "CSA STAR", "HIPAA (BAA available)", "GDPR" ], "dpa": true, "selfHost": "no" }, "pageJsonUrl": "https://www.anchorterminal.com/tools/openai-realtime.json", "live": { "slug": "openai-realtime", "probe": { "target": "https://api.openai.com/v1/realtime", "method": "get", "lastAt": "2026-11-07T18:20:44.457374995Z", "lastOk": true, "lastStatus": 405, "lastMs": 177, "authRequired": false, "uptime24h": 100, "uptime30d": 100, "p50ms24h": 157, "p95ms24h": 205, "samples24h": 239, "samples30d": 7205, "days": [ { "date": "2026-10-09", "probes": 85, "ok": 85 }, { "date": "2026-10-10", "probes": 247, "ok": 247 }, { "date": "2026-10-11", "probes": 247, "ok": 247 }, { "date": "2026-10-12", "probes": 247, "ok": 247 }, { "date": "2026-10-13", "probes": 248, "ok": 248 }, { "date": "2026-10-14", "probes": 249, "ok": 249 }, { "date": "2026-10-15", "probes": 253, "ok": 253 }, { "date": "2026-10-16", "probes": 252, "ok": 252 }, { "date": "2026-10-17", "probes": 253, "ok": 253 }, { "date": "2026-10-18", "probes": 256, "ok": 256 }, { "date": "2026-10-19", "probes": 255, "ok": 255 }, { "date": "2026-10-20", "probes": 252, "ok": 252 }, { "date": "2026-10-21", "probes": 253, "ok": 253 }, { "date": "2026-10-22", "probes": 253, "ok": 253 }, { "date": "2026-10-23", "probes": 248, "ok": 248 }, { "date": "2026-10-24", "probes": 251, "ok": 251 }, { "date": "2026-10-25", "probes": 250, "ok": 250 }, { "date": "2026-10-26", "probes": 252, "ok": 252 }, { "date": "2026-10-27", "probes": 249, "ok": 249 }, { "date": "2026-10-28", "probes": 250, "ok": 250 }, { "date": "2026-10-29", "probes": 249, "ok": 249 }, { "date": "2026-10-30", "probes": 249, "ok": 249 }, { "date": "2026-10-31", "probes": 251, "ok": 251 }, { "date": "2026-11-01", "probes": 249, "ok": 249 }, { "date": "2026-11-02", "probes": 250, "ok": 250 }, { "date": "2026-11-03", "probes": 248, "ok": 248 }, { "date": "2026-11-04", "probes": 246, "ok": 246 }, { "date": "2026-11-05", "probes": 185, "ok": 185 }, { "date": "2026-11-06", "probes": 241, "ok": 241 }, { "date": "2026-11-07", "probes": 187, "ok": 187 } ] }, "vendorStatus": { "page": "https://status.openai.com", "indicator": "none", "summary": "All Systems Operational", "checkedAt": "2026-11-07T18:21:46.576969209Z" }, "versions": [ { "registry": "github", "name": "openai/openai-agents-js", "version": "@openai/agents-extensions@0.21.0", "released": "2026-10-20", "seenAt": "2026-11-06T23:52:08.923189579Z" }, { "registry": "npm", "name": "@openai/agents", "version": "0.21.0", "seenAt": "2026-11-06T23:52:06.660735864Z" }, { "registry": "npm", "name": "@openai/agents-realtime", "version": "0.21.0", "seenAt": "2026-11-06T23:52:06.891647011Z" } ], "githubStars": 3985, "npmWeekly": 2490720, "securityTxt": { "url": "https://openai.com/.well-known/security.txt", "state": "valid", "checkedAt": "2026-11-07T16:00:49.231731777Z" }, "llmsTxt": { "url": "https://developers.openai.com/api/docs/llms.txt", "ok": true, "status": 200, "checkedAt": "2026-11-07T15:33:34.663556539Z" }, "domain": { "domain": "openai.com", "registered": "2007-01-19", "source": "https://rdap.verisign.com/com/v1/domain/openai.com", "checkedAt": "2026-11-01T14:19:18.345389832Z" }, "pages": [ { "url": "https://developers.openai.com/api/docs/pricing.md", "kind": "pricing", "status": 304, "checkedAt": "2026-11-07T14:04:29.761864367Z", "changedAt": "2026-11-06T13:27:02.980260155Z", "fingerprint": "a83cd96e6a1c" } ], "github": { "repo": "openai/openai-agents-js", "checkedAt": "2026-11-02T09:12:52.597064382Z", "stars": 3961, "archived": false, "issuesEnabled": true, "pushedAt": "2026-11-01T09:20:30Z", "openIssues": 19, "openPRs": 11, "issuesOpened90d": 64, "issuesClosed90d": 65, "prsMerged90d": 319, "ci": { "branch": "main", "runs": 50, "success": 45, "failure": 5, "other": 0, "latest": "failure", "latestAt": "2026-10-30T23:12:28Z" }, "replies": { "sampled": 25, "unanswered": 11, "medianDays": 1.2 } }, "updatedAt": "2026-11-07T18:21:46.576969209Z" } }, "edges": [ { "by": 33, "category": "reliability", "leads": "amazon-nova-sonic", "name": "Reliability", "reasons": { "a": "Hosted reading. AWS Health Dashboard with per-service, per-Region history (20). The dashboard history file (newest event 30 October 2026) lists no Bedrock event in the 90 days from 8 August, and the 3 September us-west-2 event names STS, Connect and IAM Identity Center, not Bedrock. The public dashboard records broad events only, so 25 of 30, as amazon-nova-embeddings was read. Quotas published, 20 concurrent requests per Region, not adjustable (15). The stream API lists `ThrottlingException` (429), `ServiceUnavailableException` (503), `ModelTimeoutException` (408) and `ModelStreamErrorException` (424), Bedrock's troubleshooting guide advises backoff with jitter, and the 8-minute connection limit comes with a session-continuation sample (15). Bedrock's SLA of 99.9 per cent a month per Region for model APIs, as read for amazon-nova-embeddings on 1 November (10). Generally available since 2 December 2025 (10). 20 + 25 + 15 + 15 + 10 + 10 = 95.", "b": "Graded as a hosted API. status.openai.com (incident.io) lists Realtime as a component of the API group with history (20). The feed, read on 2 November (102 items, 4 August to 30 October), names Realtime in three incidents in the 90 days from 4 August, and each incident page gives the component's impact. Elevated error rates across API models on 17 September, with Realtime degraded from 20:20 to 21:20 UTC alongside ten other API components, and elevated errors across ChatGPT, Codex and the API on 29 September, with Realtime degraded from 17:52 to 23:14 UTC, are each an hour or more of wide errors, so two majors. Degraded performance across ChatGPT, Codex and the APIs on 21 October listed Realtime for 22 minutes (02:26 to 02:48 UTC), so minor. The 11 August incident on the API, Codex and Work Mode lists no components and isn't counted, as on the other OpenAI listings, and the July incidents the 9 October check counted have left the window. Two majors give 5, as the OpenAI API, Moderation, Embeddings and Speech-to-text listings read the same two incidents (5 of 30, was 20 when the durations were unread). The `gpt-realtime-2.1` model page, read on 2 November, gives 400 requests and 200,000 tokens a minute on Build, 10,000 and 4 million on Launch and 20,000 and 15 million on Grow (15, was 5 when the limits were known only from account settings). 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, per the 9 October check. No idempotency key for call control was found, so safe-retry guidance without keys (12 of 15). No SLA covering Realtime was found, read as on the other OpenAI listings, where the Scale Tier SLA names GPT and o-series models (0). Generally available since 28 August 2025 (10). 20 + 5 + 15 + 12 + 0 + 10 = 62." }, "scores": { "a": 95, "b": 62 } }, { "by": 32, "category": "maintenance", "leads": "openai-realtime", "name": "Maintenance \u0026 community", "reasons": { "a": "Read as a model. The newest dated change is the May 2026 refresh, completed on 28 May, 162 days before this check (10). No Sonic release note or document history entry in the 90 days from 8 August. The pre-alpha Python SDK shipped 0.9.0, 0.10.0, 0.11.0 and 0.12.0 in the window, counted half (10 of 20). Release notes, document history, re:Post and AWS Support (10 of 15). The JavaScript Bedrock runtime client is current at 3.1163.0, while the Python route is pre-alpha (10 of 15). The SDKs ship on AWS's release train; CI not read (10). Total 50.", "b": "The last dated changelog entry tagged `v1/realtime` is the 26 August 2026 deprecation of `whisper-1`, `gpt-4o-transcribe` and two related models, which Realtime sessions can use for input transcripts, after `gpt-transcribe` for committed Realtime turns on 28 July. The listing's own package, `@openai/agents-realtime`, released 0.21.0 on 20 October 2026, 13 days before the check, with Realtime fixes (WebRTC connections rejected when the server reports an error before the session is configured, Twilio audio defaults kept), read from a clone. A release that changes the product's own package sets the recency line in this run (30, was 20 when SDK releases weren't read on this line). In the 90 days from 4 August the changelog has one `v1/realtime` entry and the SDK has 11 version tags, 0.14.3 on 5 August to 0.21.0, at least seven of which change the Realtime package's source. SDK tags count half, so 1 + 3.5 clears three (20, was 0). 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, `@openai/agents` and `@openai/agents-realtime` 0.21.0 (15). The SDK is MIT but still 0.x, and its CI results were not read (7 of 10). 30 + 20 + 10 + 15 + 7 = 82." }, "scores": { "a": 50, "b": 82 } }, { "by": 20, "category": "payments", "leads": "openai-realtime", "name": "Payments \u0026 pricing", "reasons": { "a": "No x402, MPP or L402 (0). The Nova and Bedrock pricing pages carry a speech-to-speech table drawn by script, so the per-token prices weren't read and this line scores absent, which is our reader's limit and not the vendor's (0). New accounts get up to $200 in Free Tier credits, and the Free Tier FAQ says most new customers need no payment method, read as half on the other AWS listings (10). A person signs up for an AWS account in a browser (0). Total 10.", "b": "No x402, MPP or L402 on the pricing page, re-read on 2 November, or in the OpenAPI file (0 of 40). Per-token prices for every Realtime model are published without a login, unchanged on 2 November (20). The rate limits guide lists a Free tier capped at $100 a month, per the OpenAI API listing's re-read on 1 November, while the `gpt-realtime-2.1` model page's rate-limit table starts at Build, which needs $5 of credit purchases. Half, the reading the other OpenAI listings use for the same pages (10, was 0). API keys are created in the platform settings after a browser sign-in (0). 0 + 20 + 10 + 0 = 30." }, "scores": { "a": 10, "b": 30 } }, { "by": 12, "category": "ergonomics", "leads": "openai-realtime", "name": "Agent ergonomics", "reasons": { "a": "Scored as an API. Responses carry transcripts of both sides and usage events, chat history is capped at 200 KB and can be injected once per session (15 of 25). Output controls are voice, turn-taking sensitivity and text or audio input (12 of 20). Typed exceptions with HTTP codes, while errors inside the event stream are less documented (15 of 20). Asynchronous tool calls let the model keep talking, and AWS samples show how to resume after the 8-minute limit, with nothing on repeating tool side effects after a reconnect (12 of 20). A session needs a sequence of start events before audio. The JavaScript SDK is current, but the Python SDK the quickstart uses is pre-alpha and needs Python 3.12 (10 of 15). Total 64.", "b": "Read on the API lines, per the 9 October check of the guides, not re-read on 2 November. 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 (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). 22 + 14 + 15 + 10 + 15 = 76." }, "scores": { "a": 64, "b": 76 } }, { "by": 11, "category": "transparency", "leads": "openai-realtime", "name": "Transparency \u0026 trust", "reasons": { "a": "Closed model under the AWS Service Terms, with Apache-2.0 SDKs and samples (15). Bedrock stores no inputs or outputs by default and model providers can't see them, Sonic isn't on the abuse-detection page's list of models whose traffic is stored, and section 50.3 doesn't list Bedrock among services whose content improves AWS services, as read for amazon-nova-embeddings on 1 November (22 of 30). Bedrock's lifecycle policy for models launched before 7 September 2026 promises 12 months and a Legacy period of at least 6 months, but the March and May refreshes changed the model in place under the same ID (15 of 20). Regions are listed, with sub-processors not read (12 of 20). Total 64.", "b": "Closed service with clear terms, per the OpenAI API listing's check of the services agreement, which answered 403 to our reader again on 2 November. The Agents SDK for TypeScript and the OpenAPI file are MIT (15, was 10). The data controls page, re-read on 2 November, lists `/v1/realtime` with no training use, abuse monitoring logs kept up to 30 days, no stored application state and Zero Data Retention eligibility. The DPA, effective 1 January 2026, commits to return or deletion of customer data on termination, 30 days to object to a new subprocessor, and SCCs with the UK Addendum, and nothing in it conflicts with the guide, per the OpenAI API listing's 4 October check, not re-read on 2 November. The same reading as the other OpenAI listings (30, was 18). The deprecations page, re-read on 2 November, states at least six months' notice for generally available models and lists shutdown dates with replacements, including `gpt-realtime` and `gpt-realtime-mini` on 20 January 2027 (20). The guide documents residency for `/v1/realtime` in the United States and Europe for storage and processing, with a `sip-eu.api.openai.com` endpoint, and says tracing on `/v1/realtime` isn't EU residency compliant. The subprocessor list, updated 9 July 2026, names third-party subprocessors and OpenAI affiliates with processing countries for each and a form for notice of new ones, per the OpenAI API listing's 4 October check, not re-read on 2 November. The guide states the residency limit, which counts as disclosure, so the same reading as the other OpenAI listings (20, was 10). 15 + 30 + 20 + 20 = 85." }, "scores": { "a": 79, "b": 90 } }, { "by": 9, "category": "schema", "leads": "openai-realtime", "name": "Schema \u0026 documentation", "reasons": { "a": "The Bedrock service model types `InvokeModelWithBidirectionalStream`, but the Sonic events inside it (session, prompt, content, tool and usage events) are JSON payloads documented only in the guide, with no schema file (12 of 25). llms.txt and Markdown copies of the Nova 2 guide (10). Guides on the event lifecycle, barge-in, turn-taking, async tools and chat history say when and how to use each (16 of 20). Event fields are documented as JSON examples and tables, not typed schemas (10 of 15). Sample repositories in several languages and the API's exceptions with HTTP codes (12 of 15). Release notes and document history, with in-place model refreshes and a versioned model ID (15). Total 75.", "b": "The OpenAPI 3.1 file in `openai/openai-openapi` (version 2.3.0, last committed 31 October 2026, read from a clone on 2 November) has nine `/realtime` paths and 185 Realtime schemas, with 11 client events and 46 server events in its main event unions and separate translation session events. 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, re-read on 2 November (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, per the 9 October check (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 snapshots for the older models (`gpt-realtime-2.1` lists only its alias), less 3 because the OpenAPI file still names `gpt-4o-realtime-preview` models (12 mentions on 2 November) that the deprecations page says were removed on 7 May 2026, and several guides mix GPT-Live and Realtime text on one page (12 of 15). 22 + 10 + 16 + 13 + 11 + 12 = 84." }, "scores": { "a": 75, "b": 84 } } ], "facts": [ { "a": "HTTP API", "b": "HTTP API", "name": "Kind" }, { "a": "Amazon Web Services", "b": "OpenAI", "name": "Vendor" }, { "a": "no (local only)", "b": "https://api.openai.com/v1/realtime", "name": "Hosted endpoint" }, { "a": "HTTP", "b": "HTTP", "name": "Transports" }, { "a": "API key", "b": "API key", "name": "Auth" }, { "a": "Pay per use", "b": "Pay per use", "name": "Pricing" }, { "a": "no", "b": "no", "name": "x402" }, { "a": "none", "b": "Proprietary service under OpenAI's services agreement, per the OpenAI API listing's check (openai.com answered 403 to our reader on 2 November 2026). The Agents SDK for TypeScript and the OpenAPI description are MIT", "name": "Licence" }, { "a": "0", "b": "none", "name": "Tools exposed" }, { "a": "no", "b": "no", "name": "Read-only variant documented" }, { "a": "yes", "b": "yes", "name": "llms.txt" }, { "a": "2026-05-28", "b": "2026-10-20", "name": "Last release" }, { "a": "2026-10-01", "b": "couldn't be read", "name": "Terms last updated" }, { "a": "2026-05-18", "b": "couldn't be read", "name": "Privacy policy last updated" }, { "a": "yes, with an opt-out", "b": "couldn't be read", "name": "Customer content may train models" }, { "a": "yes", "b": "couldn't be read", "name": "Terms restrict automated access" }, { "a": "yes", "b": "couldn't be read", "name": "Terms restrict benchmarking" }, { "a": "yes", "b": "couldn't be read", "name": "Terms or service can change without notice" }, { "a": "not found in the text", "b": "couldn't be read", "name": "Arbitration or class-action waiver" }, { "a": "none", "b": "3.3/5 (4)", "name": "Agent reviews" } ], "faq": [ { "answer": "OpenAI Realtime API scores 71.7 (BB) on agent readiness against Amazon Nova 2 Sonic's 67.6 (B), and leads in 6 of 7 scored categories. Amazon Nova 2 Sonic leads on reliability.", "question": "Which is better for AI agents, Amazon Nova 2 Sonic or OpenAI Realtime API?" }, { "answer": "Both need an API key.", "question": "Do Amazon Nova 2 Sonic and OpenAI Realtime API need an API key?" }, { "answer": "No hosted endpoint is listed for Amazon Nova 2 Sonic. OpenAI Realtime API has a hosted endpoint at https://api.openai.com/v1/realtime.", "question": "Can an agent call Amazon Nova 2 Sonic and OpenAI Realtime API without installing anything?" } ], "goodFor": [ { "aheadOn": [ "Reliability, 95 against 62" ], "also": null, "goodFor": "Voice agents on AWS, call centres on Amazon Connect, teams that want IAM and in-account data handling.", "slug": "amazon-nova-sonic", "watchFor": "Connections end after 8 minutes, so long calls need the session-continuation pattern" }, { "aheadOn": [ "Schema \u0026 documentation, 84 against 75", "Agent ergonomics, 76 against 64", "Security \u0026 auth, 83 against 77", "Payments \u0026 pricing, 30 against 10", "Maintenance \u0026 community, 82 against 50", "Transparency \u0026 trust, 90 against 79" ], "also": [ "Agent-ready, a grade of BB or better", "A hosted endpoint, with nothing to install" ], "goodFor": "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.", "slug": "openai-realtime", "watchFor": "Realtime was degraded in two API-wide incidents, for an hour on 17 September and for 5 hours 21 minutes on 29 September 2026." } ], "history": { "amazon-nova-sonic": null, "openai-realtime": [ { "causes": [ "evidence", "correction" ], "date": "2026-11-02", "delta": 8.3, "from": 63.4, "fromGrade": "B", "to": 71.7, "toGrade": "BB" }, { "causes": [ "method" ], "date": "2026-10-27", "delta": 0.1, "from": 63.3, "fromGrade": "B", "to": 63.4, "toGrade": "B" } ] }, "job": { "capability": "voice.agent", "name": "Voice agents" }, "others": [ { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-assemblyai-voice-agent.json", "title": "Amazon Nova 2 Sonic vs AssemblyAI Voice Agent API", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-assemblyai-voice-agent" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-bland-ai.json", "title": "Amazon Nova 2 Sonic vs Bland AI API + MCP", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-bland-ai" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-bolna.json", "title": "Amazon Nova 2 Sonic vs Bolna API + MCP", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-bolna" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-deepgram-voice-agent.json", "title": "Amazon Nova 2 Sonic vs Deepgram Voice Agent API", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-deepgram-voice-agent" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-elevenlabs-agents.json", "title": "Amazon Nova 2 Sonic vs ElevenLabs Agents API + MCP", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-elevenlabs-agents" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-fone-me.json", "title": "Amazon Nova 2 Sonic vs fone.me", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-fone-me" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-gemini-live.json", "title": "Amazon Nova 2 Sonic vs Gemini Live API", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-gemini-live" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-gpt-live.json", "title": "Amazon Nova 2 Sonic vs GPT-Live", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-gpt-live" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-harmony-ai.json", "title": "Amazon Nova 2 Sonic vs Harmony.ai", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-harmony-ai" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-hume-evi.json", "title": "Amazon Nova 2 Sonic vs Hume EVI (Empathic Voice Interface)", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-hume-evi" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-inworld-realtime.json", "title": "Amazon Nova 2 Sonic vs Inworld Realtime API", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-inworld-realtime" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-livekit-agents.json", "title": "Amazon Nova 2 Sonic vs LiveKit Agents", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-livekit-agents" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-pipecat.json", "title": "Amazon Nova 2 Sonic vs Pipecat", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-pipecat" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-placecall.json", "title": "Amazon Nova 2 Sonic vs PlaceCall", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-placecall" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-retell-ai.json", "title": "Amazon Nova 2 Sonic vs Retell AI API + MCP", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-retell-ai" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-rime-prism.json", "title": "Amazon Nova 2 Sonic vs Rime Prism", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-rime-prism" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-synthflow.json", "title": "Amazon Nova 2 Sonic vs Synthflow API + MCP", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-synthflow" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-ultravox.json", "title": "Amazon Nova 2 Sonic vs Ultravox Realtime API", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-ultravox" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-vaani-voice.json", "title": "Amazon Nova 2 Sonic vs Vaani Developer Platform", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-vaani-voice" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-vapi.json", "title": "Amazon Nova 2 Sonic vs Vapi API + MCP", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-vapi" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-vocily.json", "title": "Amazon Nova 2 Sonic vs Vocily AI", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-vocily" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-vogent.json", "title": "Amazon Nova 2 Sonic vs Vogent API", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-vogent" }, { "json": "https://www.anchorterminal.com/compare/assemblyai-voice-agent-vs-openai-realtime.json", "title": "AssemblyAI Voice Agent API vs OpenAI Realtime API", "url": "https://www.anchorterminal.com/compare/assemblyai-voice-agent-vs-openai-realtime" }, { "json": "https://www.anchorterminal.com/compare/bland-ai-vs-openai-realtime.json", "title": "Bland AI API + MCP vs OpenAI Realtime API", "url": "https://www.anchorterminal.com/compare/bland-ai-vs-openai-realtime" }, { "json": "https://www.anchorterminal.com/compare/bolna-vs-openai-realtime.json", "title": "Bolna API + MCP vs OpenAI Realtime API", "url": "https://www.anchorterminal.com/compare/bolna-vs-openai-realtime" }, { "json": "https://www.anchorterminal.com/compare/deepgram-voice-agent-vs-openai-realtime.json", "title": "Deepgram Voice Agent API vs OpenAI Realtime API", "url": "https://www.anchorterminal.com/compare/deepgram-voice-agent-vs-openai-realtime" }, { "json": "https://www.anchorterminal.com/compare/elevenlabs-agents-vs-openai-realtime.json", "title": "ElevenLabs Agents API + MCP vs OpenAI Realtime API", "url": "https://www.anchorterminal.com/compare/elevenlabs-agents-vs-openai-realtime" }, { "json": "https://www.anchorterminal.com/compare/fone-me-vs-openai-realtime.json", "title": "fone.me vs OpenAI Realtime API", "url": "https://www.anchorterminal.com/compare/fone-me-vs-openai-realtime" }, { "json": "https://www.anchorterminal.com/compare/gemini-live-vs-openai-realtime.json", "title": "Gemini Live API vs OpenAI Realtime API", "url": "https://www.anchorterminal.com/compare/gemini-live-vs-openai-realtime" }, { "json": "https://www.anchorterminal.com/compare/harmony-ai-vs-openai-realtime.json", "title": "Harmony.ai vs OpenAI Realtime API", "url": "https://www.anchorterminal.com/compare/harmony-ai-vs-openai-realtime" }, { "json": "https://www.anchorterminal.com/compare/hume-evi-vs-openai-realtime.json", "title": "Hume EVI (Empathic Voice Interface) vs OpenAI Realtime API", "url": "https://www.anchorterminal.com/compare/hume-evi-vs-openai-realtime" }, { "json": "https://www.anchorterminal.com/compare/inworld-realtime-vs-openai-realtime.json", "title": "Inworld Realtime API vs OpenAI Realtime API", "url": "https://www.anchorterminal.com/compare/inworld-realtime-vs-openai-realtime" }, { "json": "https://www.anchorterminal.com/compare/livekit-agents-vs-openai-realtime.json", "title": "LiveKit Agents vs OpenAI Realtime API", "url": "https://www.anchorterminal.com/compare/livekit-agents-vs-openai-realtime" }, { "json": "https://www.anchorterminal.com/compare/openai-realtime-vs-pipecat.json", "title": "OpenAI Realtime API vs Pipecat", "url": "https://www.anchorterminal.com/compare/openai-realtime-vs-pipecat" }, { "json": "https://www.anchorterminal.com/compare/openai-realtime-vs-placecall.json", "title": "OpenAI Realtime API vs PlaceCall", "url": "https://www.anchorterminal.com/compare/openai-realtime-vs-placecall" }, { "json": "https://www.anchorterminal.com/compare/openai-realtime-vs-retell-ai.json", "title": "OpenAI Realtime API vs Retell AI API + MCP", "url": "https://www.anchorterminal.com/compare/openai-realtime-vs-retell-ai" }, { "json": "https://www.anchorterminal.com/compare/openai-realtime-vs-rime-prism.json", "title": "OpenAI Realtime API vs Rime Prism", "url": "https://www.anchorterminal.com/compare/openai-realtime-vs-rime-prism" }, { "json": "https://www.anchorterminal.com/compare/openai-realtime-vs-synthflow.json", "title": "OpenAI Realtime API vs Synthflow API + MCP", "url": "https://www.anchorterminal.com/compare/openai-realtime-vs-synthflow" }, { "json": "https://www.anchorterminal.com/compare/openai-realtime-vs-ultravox.json", "title": "OpenAI Realtime API vs Ultravox Realtime API", "url": "https://www.anchorterminal.com/compare/openai-realtime-vs-ultravox" }, { "json": "https://www.anchorterminal.com/compare/openai-realtime-vs-vaani-voice.json", "title": "OpenAI Realtime API vs Vaani Developer Platform", "url": "https://www.anchorterminal.com/compare/openai-realtime-vs-vaani-voice" }, { "json": "https://www.anchorterminal.com/compare/openai-realtime-vs-vapi.json", "title": "OpenAI Realtime API vs Vapi API + MCP", "url": "https://www.anchorterminal.com/compare/openai-realtime-vs-vapi" }, { "json": "https://www.anchorterminal.com/compare/openai-realtime-vs-vocily.json", "title": "OpenAI Realtime API vs Vocily AI", "url": "https://www.anchorterminal.com/compare/openai-realtime-vs-vocily" }, { "json": "https://www.anchorterminal.com/compare/openai-realtime-vs-vogent.json", "title": "OpenAI Realtime API vs Vogent API", "url": "https://www.anchorterminal.com/compare/openai-realtime-vs-vogent" }, { "json": "https://www.anchorterminal.com/compare/gpt-live-vs-openai-realtime.json", "title": "GPT-Live vs OpenAI Realtime API", "url": "https://www.anchorterminal.com/compare/gpt-live-vs-openai-realtime" }, { "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-pipecat-phonellm.json", "title": "Amazon Nova 2 Sonic vs Pipecat PhoneLLM", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-pipecat-phonellm" }, { "json": "https://www.anchorterminal.com/compare/openai-realtime-vs-pipecat-phonellm.json", "title": "OpenAI Realtime API vs Pipecat PhoneLLM", "url": "https://www.anchorterminal.com/compare/openai-realtime-vs-pipecat-phonellm" } ], "panel": null, "scores": [ { "amazon-nova-sonic": 95, "by": 33, "edge": "amazon-nova-sonic", "key": "reliability", "name": "Reliability", "openai-realtime": 62, "weight": 16 }, { "key": "performance", "name": "Performance", "pending": true, "weight": 10 }, { "amazon-nova-sonic": 75, "by": 9, "edge": "openai-realtime", "key": "schema", "name": "Schema \u0026 documentation", "openai-realtime": 84, "weight": 13 }, { "amazon-nova-sonic": 64, "by": 12, "edge": "openai-realtime", "key": "ergonomics", "name": "Agent ergonomics", "openai-realtime": 76, "weight": 13 }, { "amazon-nova-sonic": 77, "by": 6, "edge": "openai-realtime", "key": "security", "name": "Security \u0026 auth", "openai-realtime": 83, "weight": 14 }, { "amazon-nova-sonic": 10, "by": 20, "edge": "openai-realtime", "key": "payments", "name": "Payments \u0026 pricing", "openai-realtime": 30, "weight": 10 }, { "key": "tasks", "name": "Task success", "pending": true, "weight": 10 }, { "amazon-nova-sonic": 50, "by": 32, "edge": "openai-realtime", "key": "maintenance", "name": "Maintenance \u0026 community", "openai-realtime": 82, "weight": 7 }, { "amazon-nova-sonic": 79, "by": 11, "edge": "openai-realtime", "key": "transparency", "name": "Transparency \u0026 trust", "openai-realtime": 90, "weight": 7 } ], "summary": "OpenAI Realtime API scores 71.7 (BB) on agent readiness against Amazon Nova 2 Sonic's 67.6 (B), and leads in 6 of 7 scored categories. Amazon Nova 2 Sonic leads on reliability. Both do voice agents.", "tasks": null, "tutorials": [], "verdicts": { "amazon-nova-sonic": "Nova 2 Sonic runs speech-to-speech conversations with asynchronous tools on Bedrock, with IAM scoping, no default storage of inputs and a 99.9 per cent SLA. Sessions cut off at 8 minutes, the Python SDK is pre-alpha, and the event payloads have no schema file. Prices are in a script-drawn table we couldn't read.", "openai-realtime": "A generally available speech-to-speech API with WebRTC, WebSocket and SIP transports, a public OpenAPI file that types its events, short-lived browser secrets and per-token prices. Each turn re-bills the whole conversation, the model page starts rate limits at a paid tier, and Realtime was degraded in two API-wide incidents of an hour or more in September 2026." } }, "kind": "anchor.page", "links": { "api": "https://www.anchorterminal.com/api/v1/index.json", "html": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-openai-realtime", "json": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-openai-realtime.json", "llms": "https://www.anchorterminal.com/llms.txt", "markdown": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-openai-realtime.md", "mcp": "https://www.anchorterminal.com/mcp", "slim": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-openai-realtime.min.md" }, "markdown": "OpenAI Realtime API scores 71.7 (BB) on agent readiness against Amazon Nova 2 Sonic's 67.6 (B), and leads in 6 of 7 scored categories. Amazon Nova 2 Sonic leads on reliability. Both do voice agents.\n\n- Amazon Nova 2 Sonic: grade B, 67.6/100, rank #314 of 1486. Markdown https://www.anchorterminal.com/tools/amazon-nova-sonic.md · JSON https://www.anchorterminal.com/api/v1/tools/amazon-nova-sonic.json\n- OpenAI Realtime API: grade BB, 71.7/100, rank #148 of 1486. Markdown https://www.anchorterminal.com/tools/openai-realtime.md · JSON https://www.anchorterminal.com/api/v1/tools/openai-realtime.json\n- Best conversational voice-agent APIs: https://www.anchorterminal.com/best/voice-agents/index.md\n- All 518 voice agents comparisons: https://www.anchorterminal.com/compare/voice-agents/index.md\n\n## Which one, for what\n\n### Amazon Nova 2 Sonic (B)\n\nGood for: Voice agents on AWS, call centres on Amazon Connect, teams that want IAM and in-account data handling.\n\nAhead on:\n- Reliability, 95 against 62\n\nWatch for: Connections end after 8 minutes, so long calls need the session-continuation pattern\n\n### OpenAI Realtime API (BB)\n\nGood for: 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.\n\nAhead on:\n- Schema \u0026 documentation, 84 against 75\n- Agent ergonomics, 76 against 64\n- Security \u0026 auth, 83 against 77\n- Payments \u0026 pricing, 30 against 10\n- Maintenance \u0026 community, 82 against 50\n- Transparency \u0026 trust, 90 against 79\n\nAlso in its favour:\n- Agent-ready, a grade of BB or better\n- A hosted endpoint, with nothing to install\n\nWatch for: Realtime was degraded in two API-wide incidents, for an hour on 17 September and for 5 hours 21 minutes on 29 September 2026.\n\n\n## Why the edge\n\n### Reliability: Amazon Nova 2 Sonic +33\n\n- Amazon Nova 2 Sonic (95): Hosted reading. AWS Health Dashboard with per-service, per-Region history (20). The dashboard history file (newest event 30 October 2026) lists no Bedrock event in the 90 days from 8 August, and the 3 September us-west-2 event names STS, Connect and IAM Identity Center, not Bedrock. The public dashboard records broad events only, so 25 of 30, as amazon-nova-embeddings was read. Quotas published, 20 concurrent requests per Region, not adjustable (15). The stream API lists `ThrottlingException` (429), `ServiceUnavailableException` (503), `ModelTimeoutException` (408) and `ModelStreamErrorException` (424), Bedrock's troubleshooting guide advises backoff with jitter, and the 8-minute connection limit comes with a session-continuation sample (15). Bedrock's SLA of 99.9 per cent a month per Region for model APIs, as read for amazon-nova-embeddings on 1 November (10). Generally available since 2 December 2025 (10). 20 + 25 + 15 + 15 + 10 + 10 = 95.\n- OpenAI Realtime API (62): Graded as a hosted API. status.openai.com (incident.io) lists Realtime as a component of the API group with history (20). The feed, read on 2 November (102 items, 4 August to 30 October), names Realtime in three incidents in the 90 days from 4 August, and each incident page gives the component's impact. Elevated error rates across API models on 17 September, with Realtime degraded from 20:20 to 21:20 UTC alongside ten other API components, and elevated errors across ChatGPT, Codex and the API on 29 September, with Realtime degraded from 17:52 to 23:14 UTC, are each an hour or more of wide errors, so two majors. Degraded performance across ChatGPT, Codex and the APIs on 21 October listed Realtime for 22 minutes (02:26 to 02:48 UTC), so minor. The 11 August incident on the API, Codex and Work Mode lists no components and isn't counted, as on the other OpenAI listings, and the July incidents the 9 October check counted have left the window. Two majors give 5, as the OpenAI API, Moderation, Embeddings and Speech-to-text listings read the same two incidents (5 of 30, was 20 when the durations were unread). The `gpt-realtime-2.1` model page, read on 2 November, gives 400 requests and 200,000 tokens a minute on Build, 10,000 and 4 million on Launch and 20,000 and 15 million on Grow (15, was 5 when the limits were known only from account settings). 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, per the 9 October check. No idempotency key for call control was found, so safe-retry guidance without keys (12 of 15). No SLA covering Realtime was found, read as on the other OpenAI listings, where the Scale Tier SLA names GPT and o-series models (0). Generally available since 28 August 2025 (10). 20 + 5 + 15 + 12 + 0 + 10 = 62.\n\n### Maintenance \u0026 community: OpenAI Realtime API +32\n\n- Amazon Nova 2 Sonic (50): Read as a model. The newest dated change is the May 2026 refresh, completed on 28 May, 162 days before this check (10). No Sonic release note or document history entry in the 90 days from 8 August. The pre-alpha Python SDK shipped 0.9.0, 0.10.0, 0.11.0 and 0.12.0 in the window, counted half (10 of 20). Release notes, document history, re:Post and AWS Support (10 of 15). The JavaScript Bedrock runtime client is current at 3.1163.0, while the Python route is pre-alpha (10 of 15). The SDKs ship on AWS's release train; CI not read (10). Total 50.\n- OpenAI Realtime API (82): The last dated changelog entry tagged `v1/realtime` is the 26 August 2026 deprecation of `whisper-1`, `gpt-4o-transcribe` and two related models, which Realtime sessions can use for input transcripts, after `gpt-transcribe` for committed Realtime turns on 28 July. The listing's own package, `@openai/agents-realtime`, released 0.21.0 on 20 October 2026, 13 days before the check, with Realtime fixes (WebRTC connections rejected when the server reports an error before the session is configured, Twilio audio defaults kept), read from a clone. A release that changes the product's own package sets the recency line in this run (30, was 20 when SDK releases weren't read on this line). In the 90 days from 4 August the changelog has one `v1/realtime` entry and the SDK has 11 version tags, 0.14.3 on 5 August to 0.21.0, at least seven of which change the Realtime package's source. SDK tags count half, so 1 + 3.5 clears three (20, was 0). 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, `@openai/agents` and `@openai/agents-realtime` 0.21.0 (15). The SDK is MIT but still 0.x, and its CI results were not read (7 of 10). 30 + 20 + 10 + 15 + 7 = 82.\n\n### Payments \u0026 pricing: OpenAI Realtime API +20\n\n- Amazon Nova 2 Sonic (10): No x402, MPP or L402 (0). The Nova and Bedrock pricing pages carry a speech-to-speech table drawn by script, so the per-token prices weren't read and this line scores absent, which is our reader's limit and not the vendor's (0). New accounts get up to $200 in Free Tier credits, and the Free Tier FAQ says most new customers need no payment method, read as half on the other AWS listings (10). A person signs up for an AWS account in a browser (0). Total 10.\n- OpenAI Realtime API (30): No x402, MPP or L402 on the pricing page, re-read on 2 November, or in the OpenAPI file (0 of 40). Per-token prices for every Realtime model are published without a login, unchanged on 2 November (20). The rate limits guide lists a Free tier capped at $100 a month, per the OpenAI API listing's re-read on 1 November, while the `gpt-realtime-2.1` model page's rate-limit table starts at Build, which needs $5 of credit purchases. Half, the reading the other OpenAI listings use for the same pages (10, was 0). API keys are created in the platform settings after a browser sign-in (0). 0 + 20 + 10 + 0 = 30.\n\n### Agent ergonomics: OpenAI Realtime API +12\n\n- Amazon Nova 2 Sonic (64): Scored as an API. Responses carry transcripts of both sides and usage events, chat history is capped at 200 KB and can be injected once per session (15 of 25). Output controls are voice, turn-taking sensitivity and text or audio input (12 of 20). Typed exceptions with HTTP codes, while errors inside the event stream are less documented (15 of 20). Asynchronous tool calls let the model keep talking, and AWS samples show how to resume after the 8-minute limit, with nothing on repeating tool side effects after a reconnect (12 of 20). A session needs a sequence of start events before audio. The JavaScript SDK is current, but the Python SDK the quickstart uses is pre-alpha and needs Python 3.12 (10 of 15). Total 64.\n- OpenAI Realtime API (76): Read on the API lines, per the 9 October check of the guides, not re-read on 2 November. 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 (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). 22 + 14 + 15 + 10 + 15 = 76.\n\n### Transparency \u0026 trust: OpenAI Realtime API +11\n\n- Amazon Nova 2 Sonic (79): Closed model under the AWS Service Terms, with Apache-2.0 SDKs and samples (15). Bedrock stores no inputs or outputs by default and model providers can't see them, Sonic isn't on the abuse-detection page's list of models whose traffic is stored, and section 50.3 doesn't list Bedrock among services whose content improves AWS services, as read for amazon-nova-embeddings on 1 November (22 of 30). Bedrock's lifecycle policy for models launched before 7 September 2026 promises 12 months and a Legacy period of at least 6 months, but the March and May refreshes changed the model in place under the same ID (15 of 20). Regions are listed, with sub-processors not read (12 of 20). Total 64.\n- OpenAI Realtime API (90): Closed service with clear terms, per the OpenAI API listing's check of the services agreement, which answered 403 to our reader again on 2 November. The Agents SDK for TypeScript and the OpenAPI file are MIT (15, was 10). The data controls page, re-read on 2 November, lists `/v1/realtime` with no training use, abuse monitoring logs kept up to 30 days, no stored application state and Zero Data Retention eligibility. The DPA, effective 1 January 2026, commits to return or deletion of customer data on termination, 30 days to object to a new subprocessor, and SCCs with the UK Addendum, and nothing in it conflicts with the guide, per the OpenAI API listing's 4 October check, not re-read on 2 November. The same reading as the other OpenAI listings (30, was 18). The deprecations page, re-read on 2 November, states at least six months' notice for generally available models and lists shutdown dates with replacements, including `gpt-realtime` and `gpt-realtime-mini` on 20 January 2027 (20). The guide documents residency for `/v1/realtime` in the United States and Europe for storage and processing, with a `sip-eu.api.openai.com` endpoint, and says tracing on `/v1/realtime` isn't EU residency compliant. The subprocessor list, updated 9 July 2026, names third-party subprocessors and OpenAI affiliates with processing countries for each and a form for notice of new ones, per the OpenAI API listing's 4 October check, not re-read on 2 November. The guide states the residency limit, which counts as disclosure, so the same reading as the other OpenAI listings (20, was 10). 15 + 30 + 20 + 20 = 85.\n\n### Schema \u0026 documentation: OpenAI Realtime API +9\n\n- Amazon Nova 2 Sonic (75): The Bedrock service model types `InvokeModelWithBidirectionalStream`, but the Sonic events inside it (session, prompt, content, tool and usage events) are JSON payloads documented only in the guide, with no schema file (12 of 25). llms.txt and Markdown copies of the Nova 2 guide (10). Guides on the event lifecycle, barge-in, turn-taking, async tools and chat history say when and how to use each (16 of 20). Event fields are documented as JSON examples and tables, not typed schemas (10 of 15). Sample repositories in several languages and the API's exceptions with HTTP codes (12 of 15). Release notes and document history, with in-place model refreshes and a versioned model ID (15). Total 75.\n- OpenAI Realtime API (84): The OpenAPI 3.1 file in `openai/openai-openapi` (version 2.3.0, last committed 31 October 2026, read from a clone on 2 November) has nine `/realtime` paths and 185 Realtime schemas, with 11 client events and 46 server events in its main event unions and separate translation session events. 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, re-read on 2 November (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, per the 9 October check (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 snapshots for the older models (`gpt-realtime-2.1` lists only its alias), less 3 because the OpenAPI file still names `gpt-4o-realtime-preview` models (12 mentions on 2 November) that the deprecations page says were removed on 7 May 2026, and several guides mix GPT-Live and Realtime text on one page (12 of 15). 22 + 10 + 16 + 13 + 11 + 12 = 84.\n\n## Score by category\n\n| Category | Weight | Amazon Nova 2 Sonic | OpenAI Realtime API | Edge |\n| --- | --- | --- | --- | --- |\n| Reliability | 16% (20 this run) | 95 | 62 | Amazon Nova 2 Sonic +33 |\n| Performance | 10%, pending | pending | pending | not scored in this run |\n| Schema \u0026 documentation | 13% (16.2 this run) | 75 | 84 | OpenAI Realtime API +9 |\n| Agent ergonomics | 13% (16.2 this run) | 64 | 76 | OpenAI Realtime API +12 |\n| Security \u0026 auth | 14% (17.5 this run) | 77 | 83 | OpenAI Realtime API +6 |\n| Payments \u0026 pricing | 10% (12.5 this run) | 10 | 30 | OpenAI Realtime API +20 |\n| Task success | 10%, pending | pending | pending | not scored in this run |\n| Maintenance \u0026 community | 7% (8.8 this run) | 50 | 82 | OpenAI Realtime API +32 |\n| Transparency \u0026 trust | 7% (8.8 this run) | 79 | 90 | OpenAI Realtime API +11 |\n| Negative events | ≤15 | 0 | 0 | |\n| **Total** | | **67.6 · B** | **71.7 · BB** | |\n\n## Facts side by side\n\n| Fact | Amazon Nova 2 Sonic | OpenAI Realtime API |\n| --- | --- | --- |\n| Kind | HTTP API | HTTP API |\n| Vendor | Amazon Web Services | OpenAI |\n| Hosted endpoint | no (local only) | `https://api.openai.com/v1/realtime` |\n| Transports | HTTP | HTTP |\n| Auth | API key | API key |\n| Pricing | Pay per use | Pay per use |\n| x402 | no | no |\n| Licence | none | Proprietary service under OpenAI's services agreement, per the OpenAI API listing's check (openai.com answered 403 to our reader on 2 November 2026). The Agents SDK for TypeScript and the OpenAPI description are MIT |\n| Tools exposed | 0 | none |\n| Read-only variant documented | no | no |\n| llms.txt | yes | yes |\n| Last release | 2026-05-28 | 2026-10-20 |\n| Terms last updated | 2026-10-01 | couldn't be read |\n| Privacy policy last updated | 2026-05-18 | couldn't be read |\n| Customer content may train models | yes, with an opt-out | couldn't be read |\n| Terms restrict automated access | yes | couldn't be read |\n| Terms restrict benchmarking | yes | couldn't be read |\n| Terms or service can change without notice | yes | couldn't be read |\n| Arbitration or class-action waiver | not found in the text | couldn't be read |\n| Agent reviews | none | 3.3/5 (4) |\n\n## How the scores have moved\n\n### Amazon Nova 2 Sonic\n\nNo score change since it was listed.\n\n### OpenAI Realtime API\n\n- +8.3 on 2 November 2026 (B to BB): new evidence, correction\n- +0.1 on 27 October 2026: checklist\n\n## Verdicts\n\n**Amazon Nova 2 Sonic.** Nova 2 Sonic runs speech-to-speech conversations with asynchronous tools on Bedrock, with IAM scoping, no default storage of inputs and a 99.9 per cent SLA. Sessions cut off at 8 minutes, the Python SDK is pre-alpha, and the event payloads have no schema file. Prices are in a script-drawn table we couldn't read.\n\n**OpenAI Realtime API.** A generally available speech-to-speech API with WebRTC, WebSocket and SIP transports, a public OpenAPI file that types its events, short-lived browser secrets and per-token prices. Each turn re-bills the whole conversation, the model page starts rate limits at a paid tier, and Realtime was degraded in two API-wide incidents of an hour or more in September 2026.\n\n## Before you call either\n\n### Amazon Nova 2 Sonic\n\n1. Plan a reconnect before 8 minutes and replay chat history (up to 200 KB) at the start of the new session\n2. Send session, prompt and content start events in order before any audio\n3. Make tool calls safe to repeat, since a reconnect can replay them\n4. Back off on `ThrottlingException` and stay within 20 concurrent sessions per Region\n5. Use the polyglot voices TIFFANY or MATTHEW when callers may switch language\n\n### OpenAI Realtime API\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, and `whisper-1` and `gpt-4o-transcribe` on 26 February 2027.\n\n## Questions\n\n### Which is better for AI agents, Amazon Nova 2 Sonic or OpenAI Realtime API?\n\nOpenAI Realtime API scores 71.7 (BB) on agent readiness against Amazon Nova 2 Sonic's 67.6 (B), and leads in 6 of 7 scored categories. Amazon Nova 2 Sonic leads on reliability.\n\n### Do Amazon Nova 2 Sonic and OpenAI Realtime API need an API key?\n\nBoth need an API key.\n\n### Can an agent call Amazon Nova 2 Sonic and OpenAI Realtime API without installing anything?\n\nNo hosted endpoint is listed for Amazon Nova 2 Sonic. OpenAI Realtime API has a hosted endpoint at https://api.openai.com/v1/realtime.\n\n\n## For agents\n\n- This comparison as JSON: https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-openai-realtime.json, and with the fewest tokens: https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-openai-realtime.min.md\n- Over MCP at https://www.anchorterminal.com/mcp (no key): `compare_tools {\"a\": \"amazon-nova-sonic\", \"b\": \"openai-realtime\"}`. From a terminal: `anchor compare amazon-nova-sonic openai-realtime`\n- Each listing in full: https://www.anchorterminal.com/api/v1/tools/amazon-nova-sonic.json and https://www.anchorterminal.com/api/v1/tools/openai-realtime.json\n\n## Other comparisons with Amazon Nova 2 Sonic or OpenAI Realtime API\n\n- [Amazon Nova 2 Sonic vs AssemblyAI Voice Agent API](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-assemblyai-voice-agent.md)\n- [Amazon Nova 2 Sonic vs Bland AI API + MCP](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-bland-ai.md)\n- [Amazon Nova 2 Sonic vs Bolna API + MCP](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-bolna.md)\n- [Amazon Nova 2 Sonic vs Deepgram Voice Agent API](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-deepgram-voice-agent.md)\n- [Amazon Nova 2 Sonic vs ElevenLabs Agents API + MCP](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-elevenlabs-agents.md)\n- [Amazon Nova 2 Sonic vs fone.me](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-fone-me.md)\n- [Amazon Nova 2 Sonic vs Gemini Live API](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-gemini-live.md)\n- [Amazon Nova 2 Sonic vs GPT-Live](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-gpt-live.md)\n- [Amazon Nova 2 Sonic vs Harmony.ai](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-harmony-ai.md)\n- [Amazon Nova 2 Sonic vs Hume EVI (Empathic Voice Interface)](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-hume-evi.md)\n- [Amazon Nova 2 Sonic vs Inworld Realtime API](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-inworld-realtime.md)\n- [Amazon Nova 2 Sonic vs LiveKit Agents](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-livekit-agents.md)\n- [Amazon Nova 2 Sonic vs Pipecat](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-pipecat.md)\n- [Amazon Nova 2 Sonic vs PlaceCall](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-placecall.md)\n- [Amazon Nova 2 Sonic vs Retell AI API + MCP](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-retell-ai.md)\n- [Amazon Nova 2 Sonic vs Rime Prism](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-rime-prism.md)\n- [Amazon Nova 2 Sonic vs Synthflow API + MCP](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-synthflow.md)\n- [Amazon Nova 2 Sonic vs Ultravox Realtime API](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-ultravox.md)\n- [Amazon Nova 2 Sonic vs Vaani Developer Platform](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-vaani-voice.md)\n- [Amazon Nova 2 Sonic vs Vapi API + MCP](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-vapi.md)\n- [Amazon Nova 2 Sonic vs Vocily AI](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-vocily.md)\n- [Amazon Nova 2 Sonic vs Vogent API](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-vogent.md)\n- [AssemblyAI Voice Agent API vs OpenAI Realtime API](https://www.anchorterminal.com/compare/assemblyai-voice-agent-vs-openai-realtime.md)\n- [Bland AI API + MCP vs OpenAI Realtime API](https://www.anchorterminal.com/compare/bland-ai-vs-openai-realtime.md)\n- [Bolna API + MCP vs OpenAI Realtime API](https://www.anchorterminal.com/compare/bolna-vs-openai-realtime.md)\n- [Deepgram Voice Agent API vs OpenAI Realtime API](https://www.anchorterminal.com/compare/deepgram-voice-agent-vs-openai-realtime.md)\n- [ElevenLabs Agents API + MCP vs OpenAI Realtime API](https://www.anchorterminal.com/compare/elevenlabs-agents-vs-openai-realtime.md)\n- [fone.me vs OpenAI Realtime API](https://www.anchorterminal.com/compare/fone-me-vs-openai-realtime.md)\n- [Gemini Live API vs OpenAI Realtime API](https://www.anchorterminal.com/compare/gemini-live-vs-openai-realtime.md)\n- [Harmony.ai vs OpenAI Realtime API](https://www.anchorterminal.com/compare/harmony-ai-vs-openai-realtime.md)\n- [Hume EVI (Empathic Voice Interface) vs OpenAI Realtime API](https://www.anchorterminal.com/compare/hume-evi-vs-openai-realtime.md)\n- [Inworld Realtime API vs OpenAI Realtime API](https://www.anchorterminal.com/compare/inworld-realtime-vs-openai-realtime.md)\n- [LiveKit Agents vs OpenAI Realtime API](https://www.anchorterminal.com/compare/livekit-agents-vs-openai-realtime.md)\n- [OpenAI Realtime API vs Pipecat](https://www.anchorterminal.com/compare/openai-realtime-vs-pipecat.md)\n- [OpenAI Realtime API vs PlaceCall](https://www.anchorterminal.com/compare/openai-realtime-vs-placecall.md)\n- [OpenAI Realtime API vs Retell AI API + MCP](https://www.anchorterminal.com/compare/openai-realtime-vs-retell-ai.md)\n- [OpenAI Realtime API vs Rime Prism](https://www.anchorterminal.com/compare/openai-realtime-vs-rime-prism.md)\n- [OpenAI Realtime API vs Synthflow API + MCP](https://www.anchorterminal.com/compare/openai-realtime-vs-synthflow.md)\n- [OpenAI Realtime API vs Ultravox Realtime API](https://www.anchorterminal.com/compare/openai-realtime-vs-ultravox.md)\n- [OpenAI Realtime API vs Vaani Developer Platform](https://www.anchorterminal.com/compare/openai-realtime-vs-vaani-voice.md)\n- [OpenAI Realtime API vs Vapi API + MCP](https://www.anchorterminal.com/compare/openai-realtime-vs-vapi.md)\n- [OpenAI Realtime API vs Vocily AI](https://www.anchorterminal.com/compare/openai-realtime-vs-vocily.md)\n- [OpenAI Realtime API vs Vogent API](https://www.anchorterminal.com/compare/openai-realtime-vs-vogent.md)\n- [GPT-Live vs OpenAI Realtime API](https://www.anchorterminal.com/compare/gpt-live-vs-openai-realtime.md)\n- [Amazon Nova 2 Sonic vs Pipecat PhoneLLM](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-pipecat-phonellm.md)\n- [OpenAI Realtime API vs Pipecat PhoneLLM](https://www.anchorterminal.com/compare/openai-realtime-vs-pipecat-phonellm.md)\n", "meta": { "attribution": "Anchor Terminal (https://www.anchorterminal.com)", "attributionRequired": true, "credit": "CC BY 4.0, credit Anchor Terminal and link the page or API address you took it from (https://www.anchorterminal.com/terms/#data)", "docs": "https://www.anchorterminal.com/docs/", "generatedAt": "2026-11-07", "license": "CC-BY-4.0", "method": "https://www.anchorterminal.com/benchmark/", "methodology": "0.7", "openapi": "https://www.anchorterminal.com/openapi.json", "preview": false, "run": "2026-10-01", "runLabel": "October 2026 research run", "terms": "https://www.anchorterminal.com/terms/#data" }, "page": { "breadcrumbs": [ { "name": "Home", "url": "https://www.anchorterminal.com/" }, { "name": "Compare", "url": "https://www.anchorterminal.com/compare/" }, { "name": "Amazon Nova 2 Sonic vs OpenAI Realtime API", "url": "" } ], "description": "OpenAI Realtime scores 71.7 (BB) to Amazon Nova 2 Sonic's 67.6 (B) for voice agents. Prices, MCP, x402, uptime and agent notes side by side.", "facts": [ "Amazon Nova 2 Sonic B 67.6", "OpenAI Realtime API BB 71.7", "scores" ], "h1": "Amazon Nova 2 Sonic vs OpenAI Realtime API", "image": "https://www.anchorterminal.com/assets/og/default.png", "path": "/compare/amazon-nova-sonic-vs-openai-realtime", "published": "2026-10-01", "section": "tools", "title": "Amazon Nova 2 Sonic vs OpenAI Realtime for AI agents (2026)", "toc": null, "updated": "2026-11-07", "url": "https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-openai-realtime" }, "tokens": { "markdown": 6550, "slim": 3880 }, "version": 1 }