{ "data": { "a": { "slug": "inworld-realtime", "name": "Inworld Realtime API", "vendor": "Inworld AI", "vendorUrl": "https://inworld.ai", "kind": "http-api", "category": "voice-agents", "summary": "Inworld's Realtime API runs spoken conversations over WebSocket or WebRTC, chaining Inworld STT, an LLM chosen by the developer and Inworld TTS, with turn detection, interruptions and tool calling. It follows OpenAI's Realtime protocol.", "url": "https://www.anchorterminal.com/tools/inworld-realtime", "markdownUrl": "https://www.anchorterminal.com/tools/inworld-realtime.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/inworld-realtime.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/inworld-realtime.json", "dns": "inworld-realtime.grade.anchorterminal.com", "transports": [ "websocket", "http" ], "packages": [ { "registry": "npm", "name": "@inworld/cli" } ], "auth": "api-key", "authNotes": "Servers send `Authorization: Basic` with the Base64 `key_id:secret` credential. Browsers send `Authorization: Bearer` with a session token minted server side. The `key` query parameter in the WebSocket URL is a session ID, not a secret.", "pricing": "usage", "pricingNotes": "No per-minute or per-session price. A session is billed as its STT ($0.10 to $0.15 an hour), its LLM tokens at the model's rate and its TTS characters (TTS-2 from $25 per million on the free plan). Segments already sent to TTS are billed after an interruption.", "priceSummary": "Pay per use", "where": "local", "x402": { "level": "no", "evidence": "No x402, MPP or L402 in the docs, llms.txt or pricing page (checked 2026-11-06). Plans and credits are bought in the Portal.", "endpoints": [] }, "toolCount": 0, "popularity": { "githubStars": null, "npmWeekly": null, "pypiWeekly": null, "asOf": "2026-11-06" }, "docsUrl": "https://docs.inworld.ai/realtime/overview", "llmsTxt": "https://docs.inworld.ai/llms.txt", "capabilities": [ "voice.agent", "voice.pipeline", "voice.tools", "voice.telephony" ], "tags": [ "hosted", "usage", "free-tier", "closed-source", "llms-txt", "streaming", "websocket", "webrtc", "openai-compatible", "enterprise" ], "lastRelease": "2026-10-16", "graded": true, "anchor": { "graded": true, "score": 60.1, "grade": "C", "agentReady": false, "rank": 668, "ranked": true, "rankOf": 1486, "categoryRank": 12, "methodology": "0.7", "run": "2026-10-01", "scores": { "ergonomics": 68, "maintenance": 71, "payments": 40, "reliability": 62, "schema": 60, "security": 58, "transparency": 63 }, "pending": [ "performance", "tasks" ], "assessment": { "confidence": "medium", "date": "2026-11-06" }, "negative": 0, "verdict": "Inworld's Realtime API speaks OpenAI's Realtime protocol over WebSocket or WebRTC, with a choice of LLM, client-side tools and dated release notes. Cost is the sum of STT, LLM and TTS usage with no per-minute price, there is no machine-readable event schema, and status history couldn't be read.", "bestFor": "Teams moving off OpenAI's Realtime API who want a choice of LLM and Inworld voices.", "strengths": [ "Follows OpenAI's Realtime events, with a migration guide that changes only the URL, key and a few defaults", "Dated release notes name affected clients and required actions for each compatibility change", "A choice of LLM through Inworld's router, with prompt caching on Anthropic models", "Concurrent sessions from 10 on the free plan to 1,000 on Growth", "Zero data retention covers session content from the Developer plan up" ], "weaknesses": [ "No per-minute price. A session's cost depends on STT, LLM tokens and TTS characters, and interrupted segments are still billed", "No AsyncAPI or other machine-readable schema for the socket events", "Status history couldn't be read, and an SLA exists only for Enterprise", "The 21 July 2026 release changed response and item shapes for existing clients", "Zero data retention doesn't cover requests routed to third-party LLM providers" ], "agentNotes": [ "Set the auto follow-up off if your client sends its own `response.create` after a tool result, or replies double", "Read `response.done` usage as an estimate and mute input audio when nobody speaks, since STT bills every second sent", "Use `response.speak` for fixed greetings, which skips the LLM", "Treat `unsupported_field` errors as warnings and keep the session open", "Mint browser session tokens on your server and never ship the Basic credential to a client" ], "metrics": { "kind": "remote", "measured": false }, "reviewCount": 0, "avgRating": 0, "history": [ { "basis": "public evidence", "confidence": "medium", "grade": "C", "methodology": "0.7", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 60.1 } ], "editorialScores": { "ergonomics": 68, "maintenance": 71, "payments": 40, "reliability": 62, "schema": 60, "security": 58, "transparency": 51 }, "provenanceScore": 75, "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": "npm install -g @inworld/cli" }, "letme": { "capability": "https://letme.dev/voice.agent", "tool": "https://letme.dev/inworld-realtime" }, "sameCompany": [ "inworld-stt", "inworld-tts", "inworld-voice-cloning" ], "area": "voice", "unitPrices": [ { "item": "Speech recognition in a session", "unit": "audio-minute", "usd": 0.0025, "note": "STT-1 at $0.15 an hour On-Demand, $0.10 on paid plans" }, { "item": "Agent speech, TTS-2", "unit": "1m-chars", "usd": 25, "note": "free On-Demand plan; lower on paid plans" } ], "provenance": { "legalEntity": "Theai, Inc. dba Inworld AI", "domain": "inworld.ai", "domainRegistered": "", "endpointOnVendorDomain": true, "terms": "https://inworld.ai/terms", "privacy": "https://inworld.ai/privacy", "statusPage": "https://status.inworld.ai", "changelog": "https://docs.inworld.ai/release-notes/realtime", "securityTxt": "none", "checked": "2026-11-06", "notes": [ "The Terms of Service (effective 11 June 2025) are an agreement with Theai, Inc. dba Inworld AI under California law, with courts in Santa Clara. The Privacy Notice (last updated 18 January 2026) names Theai, Inc.", "inworld.ai/.well-known/security.txt returns 404. The security page links a vulnerability report form.", "status.inworld.ai runs the open-source issue-status app, which draws incidents from GitHub issues in inworld-ai/inworld-issue-status. Its history wasn't read, because GitHub's robots.txt couldn't be read from our environment.", "inworld.ai's robots.txt allows Claude agents on the paths read and sets `Content-Signal: ai-train=no, search=yes, ai-input=no`. docs.inworld.ai has no such signal.", "The domain's registration date wasn't established, because the RDAP lookup failed.", "firstReleased is left empty. The public Realtime release notes start on 21 July 2026 and no launch date was found." ], "score": 75 }, "pageJsonUrl": "https://www.anchorterminal.com/tools/inworld-realtime.json", "live": { "slug": "inworld-realtime", "vendorStatus": { "page": "https://status.inworld.ai", "indicator": "unknown", "summary": "no machine-readable status found", "checkedAt": "2026-11-07T04:47:57.03709454Z" }, "versions": [ { "registry": "npm", "name": "@inworld/cli", "version": "1.5.0", "seenAt": "2026-11-06T23:35:04.551623033Z" } ], "npmWeekly": 290, "securityTxt": { "url": "https://inworld.ai/.well-known/security.txt", "state": "none", "checkedAt": "2026-11-07T16:01:11.794771951Z" }, "llmsTxt": { "url": "https://docs.inworld.ai/llms.txt", "ok": true, "status": 200, "checkedAt": "2026-11-07T15:32:46.028273902Z" }, "pages": [ { "url": "https://docs.inworld.ai/release-notes/realtime", "kind": "changelog", "status": 200, "checkedAt": "2026-11-07T14:06:42.218731652Z", "changedAt": "0001-01-01T00:00:00Z", "fingerprint": "9ae0f1f05b69" }, { "url": "https://inworld.ai/privacy", "kind": "privacy", "status": 304, "checkedAt": "2026-11-07T14:11:39.031139211Z", "changedAt": "0001-01-01T00:00:00Z", "fingerprint": "99dc39a9de92" }, { "url": "https://inworld.ai/terms", "kind": "terms", "status": 304, "checkedAt": "2026-11-07T14:11:41.181276356Z", "changedAt": "0001-01-01T00:00:00Z", "fingerprint": "777d881cbf9b" } ], "updatedAt": "2026-11-07T16:01:11.794771951Z" } }, "answer": "Pipecat PhoneLLM and Inworld Realtime API score within a point of each other on agent readiness, 60.4 (C) and 60.1 (C). Inworld Realtime API leads on agent ergonomics, maintenance \u0026 community and transparency \u0026 trust.", "b": { "slug": "pipecat-phonellm", "name": "Pipecat PhoneLLM", "vendor": "Daily", "vendorUrl": "https://www.daily.co/blog/phonellm-alpha-2/", "kind": "model", "category": "voice-agents", "summary": "Pipecat PhoneLLM is Daily's open-weight text model for the language step of phone voice agents, with tool calling. Alpha 2, released 15 October 2026, comes as Mini (30B, on Workers AI) and Max (120B, weights and dedicated endpoints).", "url": "https://www.anchorterminal.com/tools/pipecat-phonellm", "markdownUrl": "https://www.anchorterminal.com/tools/pipecat-phonellm.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/pipecat-phonellm.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/pipecat-phonellm.json", "dns": "pipecat-phonellm.grade.anchorterminal.com", "repo": "https://huggingface.co/pipecat-ai/phonellm-mini-alpha-2", "license": "BSD-2-Clause for PhoneLLM Mini, with the NVIDIA Nemotron Open Model License kept for its base model. OpenMDW-1.1 for PhoneLLM Max, the licence of its base model. Workers AI, the hosted route graded, is a service under Cloudflare's terms", "transports": [ "http" ], "remoteUrl": "https://api.cloudflare.com/client/v4/accounts/{account_id}/ai/run/@cf/daily-co/phonellm-mini-alpha-2", "packages": [], "auth": "api-key", "authNotes": "On Workers AI, a Cloudflare API token with Workers AI permission on one account, sent as a Bearer header. Inside a Worker the `AI` binding needs no token. The weights of both sizes download from Hugging Face without an account or approval, and a self-hosted vLLM server has whatever authentication the owner adds. Baseten, Modal and AWS SageMaker each need their own account.", "pricing": "freemium", "pricingNotes": "PhoneLLM Mini costs $0.06 per million input tokens and $0.24 per million output tokens on Workers AI (5,455 and 21,818 neurons), with no cached-input price listed. Workers AI includes 10,000 neurons a day on the Free and Paid Workers plans (https://developers.cloudflare.com/workers-ai/platform/pricing/, read 17 October 2026). Max has no per-token route. It runs on dedicated GPU endpoints at Baseten or Modal, billed for compute, or on the owner's hardware (two B200 or B300 GPUs for BF16, per the card). The weights are free to download.", "priceSummary": "from $0.06 / 1M in", "where": "hosted", "x402": { "level": "no", "evidence": "Cloudflare's Machine Payments beta takes x402 on `POST /ai/run` for four models, and PhoneLLM Mini isn't one of them (https://developers.cloudflare.com/ai-gateway/features/machine-payments/, read 17 October 2026). No x402, MPP or L402 on the Daily post or the model cards.", "endpoints": [] }, "toolCount": null, "popularity": { "githubStars": null, "npmWeekly": null, "pypiWeekly": null, "asOf": "2026-10-17" }, "docsUrl": "https://huggingface.co/pipecat-ai/phonellm-mini-alpha-2", "llmsTxt": "https://developers.cloudflare.com/llms.txt", "capabilities": [ "voice.pipeline", "voice.tools", "inference.llm", "inference.open-weights" ], "tags": [ "model", "open-weights", "hosted", "alpha", "new", "free-tier", "no-card", "usage-priced", "llms-txt", "openai-compatible", "telephony", "gpu" ], "lastRelease": "2026-10-15", "graded": true, "anchor": { "graded": true, "score": 60.4, "grade": "C", "agentReady": false, "rank": 650, "ranked": true, "rankOf": 1486, "categoryRank": 11, "methodology": "0.7", "run": "2026-10-01", "scores": { "ergonomics": 61, "maintenance": 58, "payments": 40, "reliability": 58, "schema": 72, "security": 69, "transparency": 58 }, "pending": [ "performance", "tasks" ], "assessment": { "confidence": "medium", "date": "2026-11-04" }, "negative": 0, "verdict": "PhoneLLM Mini is callable per token on Workers AI at $0.06 input and $0.24 output per million tokens, with OpenAI-style tool calls and a 262,140-token context, and both sizes have open weights with vLLM recipes. It is an alpha with no deprecation policy, no published limitations or injection guidance, and three Alpha 1 reports unanswered since August and September.", "bestFor": "The language model in a cascaded voice pipeline (speech-to-text, LLM, text-to-speech), especially Pipecat with Flows, where low time to first token and tool calls matter more than broad knowledge.", "strengths": [ "Mini is on Workers AI at $0.06 per million input and $0.24 per million output tokens, inside the 10,000 free neurons a day", "Both sizes are trained for multi-turn tool calling and Pipecat Flows, and take OpenAI-style `tools` and `tool_choice`", "Ungated weights in BF16, FP8 and NVFP4, with vLLM 0.30 recipes measured on B200 GPUs in each card", "Mini is BSD-2-Clause and Max OpenMDW-1.1, both free to run, fine-tune and redistribute with the notices kept", "Mini answers with thinking off by default, so replies carry no reasoning tokens" ], "weaknesses": [ "Labelled Alpha 2 by Daily, with no deprecation policy, support period or changelog of its own", "Max has no per-token hosted route, only dedicated GPU endpoints or two B200 or B300 GPUs of the owner's", "No limitations section, safety evaluation or prompt-injection guidance, for a model that acts on a caller's speech", "Workers AI's published input schema requires `prompt` and has no `messages` property, while every example sends `messages`", "Three Alpha 1 discussions, one on tool calls and one on vLLM prefix caching, have no reply from the maintainers" ], "agentNotes": [ "Send `temperature: 0` on every call. The repositories' generation_config.json samples by default, and Max's sets temperature 1.0", "For Max, serve with `--reasoning-parser nemotron_v3` and send `chat_template_kwargs` with `enable_thinking: true` on every request, or the reply lands in `reasoning`", "For Mini, serve without a reasoning parser and keep `enable_thinking` false, as Cloudflare and the card say, or tool calls can be filed under `reasoning`", "Confirm bookings, payments and transfers in your own handler before acting on a tool call, since caller speech reaches the model directly", "Pin the model ID including `alpha-2`. Daily states no notice period, and a later alpha may replace it" ], "metrics": { "kind": "remote", "measured": false }, "reviewCount": 4, "avgRating": 2.8, "history": [ { "basis": "public evidence", "confidence": "medium", "grade": "C", "methodology": "0.7", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 60.4 } ], "editorialScores": { "ergonomics": 61, "maintenance": 58, "payments": 40, "reliability": 58, "schema": 72, "security": 69, "transparency": 55 }, "provenanceScore": 61, "checked": { "categories": { "ergonomics": "2026-11-04", "maintenance": "2026-11-04", "payments": "2026-11-04", "reliability": "2026-11-04", "schema": "2026-11-04", "security": "2026-11-04", "transparency": "2026-11-04" }, "last": "2026-11-04", "lastMoved": "2026-11-04", "oldest": "2026-11-04", "stale": false, "staleAfterDays": 90 } }, "connect": { "install": "vllm serve pipecat-ai/phonellm-mini-alpha-2 \\\n --served-model-name phonellm-mini-alpha-2 \\\n --trust-remote-code \\\n --dtype bfloat16 \\\n --max-model-len 32768 \\\n --max-num-seqs 256 \\\n --enable-prefix-caching \\\n --async-scheduling \\\n --mamba-ssm-cache-dtype float32 \\\n --enable-auto-tool-choice \\\n --tool-call-parser qwen3_coder", "http": "curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/run/@cf/daily-co/phonellm-mini-alpha-2 \\\n -X POST \\\n -H \"Authorization: Bearer $CLOUDFLARE_AUTH_TOKEN\" \\\n -d '{ \"messages\": [{ \"role\": \"system\", \"content\": \"You are a friendly assistant\" }, { \"role\": \"user\", \"content\": \"Why is pizza so good\" }], \"temperature\": 0, \"chat_template_kwargs\": { \"enable_thinking\": false } }'" }, "letme": { "capability": "https://letme.dev/voice.pipeline", "tool": "https://letme.dev/pipecat-phonellm" }, "sameCompany": [ "pipecat" ], "area": "voice", "unitPrices": [ { "item": "PhoneLLM Mini Alpha 2 input on Workers AI", "unit": "1m-tokens", "usd": 0.06, "note": "5,455 neurons. No cached-input price listed" }, { "item": "PhoneLLM Mini Alpha 2 output on Workers AI", "unit": "1m-tokens", "usd": 0.24, "note": "21,818 neurons" } ], "provenance": { "legalEntity": "Daily", "domain": "pipecat.ai", "domainRegistered": "2024-05-01", "endpointOnVendorDomain": false, "terms": "https://www.daily.co/legal/terms-of-service/", "privacy": "https://www.daily.co/legal/privacy/", "statusPage": "https://www.cloudflarestatus.com", "changelog": "", "securityTxt": "none", "checked": "2026-10-17", "notes": [ "The weights are published by the pipecat-ai organisation on Hugging Face, and Mini's LICENSE reads 'Copyright (c) 2026, Daily'. Legal entity, domain date, terms, privacy and security.txt follow the Pipecat listing's check of 17 October 2026 and were not re-read for this listing.", "The graded hosted route is Cloudflare's, at api.cloudflare.com, so the status page is Cloudflare's. Daily runs no endpoint of its own for PhoneLLM.", "No changelog of PhoneLLM's own was found. Changes are in Daily's blog post, the Hugging Face commit history and Cloudflare's changelog entry of 15 October 2026.", "Pages were read with curl after each host's robots.txt (www.daily.co, huggingface.co, developers.cloudflare.com, www.cloudflarestatus.com, www.pipecat.ai, modal.com)." ], "score": 61 }, "pageJsonUrl": "https://www.anchorterminal.com/tools/pipecat-phonellm.json", "live": { "slug": "pipecat-phonellm", "probe": { "target": "https://api.cloudflare.com/client/v4/accounts/{account_id}/ai/run/@cf/daily-co/phonellm-mini-alpha-2", "method": "get", "lastAt": "2026-11-07T18:20:47.174467272Z", "lastOk": true, "lastStatus": 404, "lastMs": 44, "authRequired": false, "uptime24h": 100, "uptime30d": 100, "p50ms24h": 21, "p95ms24h": 82, "samples24h": 239, "samples30d": 5145, "days": [ { "date": "2026-10-17", "probes": 21, "ok": 21 }, { "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://www.cloudflarestatus.com", "indicator": "major", "summary": "Partial System Outage", "checkedAt": "2026-11-07T18:21:54.314052414Z" }, "securityTxt": { "url": "https://pipecat.ai/.well-known/security.txt", "state": "none", "checkedAt": "2026-11-07T16:00:50.311611328Z" }, "llmsTxt": { "url": "https://developers.cloudflare.com/llms.txt", "ok": true, "status": 200, "checkedAt": "2026-11-07T15:33:37.648152458Z" }, "domain": { "domain": "pipecat.ai", "registered": "2024-05-01", "source": "https://rdap.identitydigital.services/rdap/domain/pipecat.ai", "checkedAt": "2026-11-01T14:20:09.941363148Z" }, "pages": [ { "url": "https://www.daily.co/legal/privacy/", "kind": "privacy", "status": 200, "checkedAt": "2026-11-07T14:27:40.57157333Z", "changedAt": "0001-01-01T00:00:00Z", "fingerprint": "0100275d4e6b" }, { "url": "https://www.daily.co/legal/terms-of-service/", "kind": "terms", "status": 200, "checkedAt": "2026-11-07T14:27:43.567822095Z", "changedAt": "0001-01-01T00:00:00Z", "fingerprint": "c64f8e3a735a" } ], "updatedAt": "2026-11-07T18:21:54.314052414Z" } }, "edges": [ { "by": 13, "category": "maintenance", "leads": "inworld-realtime", "name": "Maintenance \u0026 community", "reasons": { "a": "Release note of 16 October 2026 (`cache_transcript`), 21 days ago (30). Three dated Realtime entries in the 90 days from 8 August, on 14 September, 23 September and 16 October (20). Discord and support@inworld.ai, with no evidence of answers read (8 of 15). Works with OpenAI's Realtime clients and the CLI is current at 1.5.0, but there is no Inworld Realtime SDK (8 of 15). CLI on npm under MIT, CI not read (5 of 10). Total 71.", "b": "Model reading, with notice periods and churn in place of release counts. Alpha 2 released on 15 October 2026, and no newer PhoneLLM repository or file change is on Hugging Face on 4 November (30). Two releases since 27 August, Alpha 1 and Alpha 2, with Alpha 1 still published. Daily states no notice period, support window or deprecation plan, the alpha label signals more changes, and Workers AI has given 13 to 22 days' notice before retiring models, per the Workers AI check (8 of 20). Three Alpha 1 discussions opened on 31 August and 16 September, a vLLM caching bug, a request for FP8 weights and a tool-call fault, still have no reply on 4 November (one post each, counted by script), though FP8 weights did ship with Alpha 2. Alpha 2's two repositories have no discussions. Daily's Discord is linked from the post and wasn't sampled (4 of 25). vLLM 0.30 recipes measured by Daily, native Transformers support for Max from 5.8, and Workers AI with Cloudflare's SDKs. Mini still needs repository code (12 of 15). Max's generation config names Transformers 5.8.1 and Mini's 4.57.6. No CI or tests for the weights visible to us (4 of 10). 58." }, "scores": { "a": 71, "b": 58 } }, { "by": 12, "category": "schema", "leads": "pipecat-phonellm", "name": "Schema \u0026 documentation", "reasons": { "a": "No AsyncAPI or other machine-readable file for the socket, and the spec paths redirect (0). llms.txt, Markdown twins and a docs MCP server (10). The overview, OpenAI migration guide, `providerData` reference and the billing page's account of how a session is counted say what to use when (16 of 20). The reference lists every client and server event with field types and required marks, as text rather than a schema (12 of 15). JavaScript and Python example projects and an error event with `type`, `code` and `param`, without a list of codes (9 of 15). Dated release notes with affected clients and required actions, but no API version beyond `/v1` (13 of 15). Total 60.", "b": "Model reading on the API lines. Each Workers AI model page links JSON Schemas for input and output. Mini's synchronous input schema lists `required: [\"prompt\"]` and has no `messages` property, while the page's TypeScript, Python and curl examples and Cloudflare's changelog all send `messages`. It also carries `audio`, `modalities` and `web_search_options`, which a text-only model has no use for (18 of 25). developers.cloudflare.com/llms.txt and a Markdown copy of the model page and changelog. The cards are Markdown on Hugging Face (10). The post and cards say what each size is for (Mini for speed, Max for accuracy, both for phone agents with tool calling and Pipecat Flows). Neither card has a limitations or out-of-scope section, and English-only is in the metadata, not the text (12 of 20). `temperature` bounded 0 to 2, `tool_choice` an enum or a forced function, `response_format` a typed one-of, and `chat_template_kwargs.enable_thinking` typed and described (12 of 15). Examples in TypeScript, Python, curl and vLLM recipes for both sizes. The Workers AI examples omit the `temperature: 0` and `enable_thinking: false` that Cloudflare's changelog asks for, and the error table is Workers AI's generic one (11 of 15). The version is in the model ID (`alpha-2`), Alpha 1 stays published, and Hugging Face keeps commit history (three commits each). No changelog of PhoneLLM's own. Changes are listed in the post and Cloudflare's changelog entry (9 of 15). 72." }, "scores": { "a": 60, "b": 72 } }, { "by": 11, "category": "security", "leads": "pipecat-phonellm", "name": "Security \u0026 auth", "reasons": { "a": "Workspace API keys sent as Basic credentials, deleted to revoke, with separate keys per environment advised. The CLI mints read-only keys unless `--write` is given, but write toggles cover only the Voice and Router APIs, so no key is limited to one product. Browsers use short-lived tokens minted server side. Session IDs, not secrets, go in the URL (25 of 30). Tools run in the client, so the developer approves side effects, with no approval step in the API (10 of 20). The agent listens to callers and passes their speech to an LLM, and no prompt-injection guidance was found (5 of 15). Usage per key and a Performance page with errors and throttling (8 of 15). SOC 2 Type II and HIPAA claimed and a vulnerability report form, with no security.txt or bug bounty, per the 4 November reading (10 of 20). Total 58. Zero data retention covers session content from the Developer plan up, but not requests routed to third-party LLM providers.", "b": "Model reading. On Workers AI, Cloudflare API tokens limited to Workers AI permissions on chosen accounts, revocable, with expiry and IP filters, per the Workers AI listing's checks. Permission is per account, not per model (27 of 30). Cloudflare's data usage page and service terms say customer content isn't used to train models without consent, with clause 9 allowing use to run and improve services, per the same checks. The model page now marks Zero data retention as Yes, which we haven't matched to a definition page, so the line holds. Self-hosted, prompts stay on the owner's hardware (17 of 20). The model is built to take a caller's transcribed speech and call tools. Neither card nor the post has limitations, a safety evaluation, or guidance on injected or hostile input. The template wraps tool results in `\u003ctool_response\u003e` and does nothing to neutralise caller text. Mini self-hosted runs modelling code from the repository with `--trust-remote-code` (3 of 15). Direct Workers AI calls have no per-request log that we found. AI Gateway logs prompts and responses when a call is routed through it (6 of 15). For the hosted route, Cloudflare's security.txt names HackerOne, and its ISO page lists ISO 27001:2022, 27018 and 27701 for the global cloud platform, which the Cloudflare listings read at 18 in this run. Daily has no security.txt and a disclosure policy with no bounty, per the Pipecat check, and the cards give no security contact, so 2 held back from Cloudflare's value (16 of 20). 27 + 17 + 3 + 6 + 16 = 69." }, "scores": { "a": 58, "b": 69 } }, { "by": 7, "category": "ergonomics", "leads": "inworld-realtime", "name": "Agent ergonomics", "reasons": { "a": "Scored as an API. Conversation history defaults to the last 100 items, can be trimmed, and can be cached on Anthropic models, and every `response.done` carries a `usage` block (18 of 25). No lists to page. Output controls are modalities, turn detection settings, optional transcription and `response.speak` (12 of 20). Error events follow OpenAI's `type` and `code` shape, unsupported fields come back as non-fatal `unsupported_field` warnings, and `item_not_found` names the bad parameter (14 of 20). Tools run on the client and a follow-up response starts automatically after a tool result. Nothing covers side effects of a tool call cut off by an interruption (12 of 20). Works with OpenAI Realtime clients after changing the URL and key, with no Inworld Realtime SDK (12 of 15). Total 68.", "b": "Model reading (tool use, structured output, caching, context, batch, SDKs, errors and defaults). Trained for multi-turn tool calls and Pipecat Flows, with OpenAI-style `tools`, `tool_choice` and `parallel_tool_calls` on Workers AI. Daily's PhoneBench gives 5.9% tool-call failures for Mini and 3.1% for Max (vendor-reported). Alpha 1 discussion 3 reports the model confirming details back in place of calling a tool, unanswered, and we don't know whether Alpha 2 fixes it (15 of 20). `json_object` and `json_schema` in the Workers AI schema. The Workers AI docs say adherence isn't guaranteed, and neither card mentions structured output (8 of 15). The vLLM recipes turn on prefix caching, but Alpha 1 discussion 1 reports it as a no-op for nemotron_h in vLLM, unanswered. Workers AI lists no cached price for Mini (5 of 15). 262,140 tokens on Workers AI and 262,144 in Mini's config, 1,048,576 in Max's config. The cards' recipes cap at 32,768 (13 of 15). The Mini model page has no Batch tag, so the Workers AI batch route isn't open to it. Self-hosted, the cards suggest 128 to 256 requests in flight per GPU (3 of 10). Any OpenAI SDK against `/v1/chat/completions`, the Workers binding, Cloudflare's TypeScript, Python and Go SDKs, and Pipecat's OpenAI-compatible services (9 of 10). Workers AI errors carry internal codes. Defaults work against the caller. Both repositories' generation_config.json sample by default (Max at temperature 1.0) where Daily recommends 0, and Max needs `chat_template_kwargs` on every request or its reply lands in `reasoning` (8 of 15). 61." }, "scores": { "a": 68, "b": 61 } }, { "by": 5, "category": "transparency", "leads": "inworld-realtime", "name": "Transparency \u0026 trust", "reasons": { "a": "Closed service with clear terms naming Theai, Inc. dba Inworld AI (15). The ZDR page says what a session keeps and that third-party LLM requests aren't covered, but the terms and privacy notice disagree on training (12 of 30). Release notes flag potentially breaking changes with the clients affected and what to do, without stating a notice period (14 of 20). US headquarters with EU and India residency on Enterprise, and no sub-processor list in the docs, while sessions may route to third-party LLM providers (10 of 20). Total 51.", "b": "Mini is BSD-2-Clause, an OSI licence, with NVIDIA's Nemotron Open Model License kept for the base work, and the card explains how the two apply. Max is OpenMDW-1.1, permissive but not on the OSI list. Training data and code aren't published, and the hosted route runs under Cloudflare's terms (22 of 30). The post says training data is fully synthetic, from unnamed open-weight teachers. On Workers AI, Cloudflare's no-training statements agree with each other apart from clause 9, per the Workers AI check. Self-hosted, nothing leaves the owner's hardware. Daily's privacy policy and DPA weren't re-read for the models (18 of 30). No deprecation policy or dated notices from Daily. Workers AI posts dated notices but has no policy (5 of 20). Cloudflare publishes a sub-processor page, and where a Workers AI request runs isn't stated beyond its network. The weights are safetensors, but Mini's repository ships Python modelling code that a self-hosted run executes (10 of 20). 55." }, "scores": { "a": 63, "b": 58 } } ], "facts": [ { "a": "HTTP API", "b": "Model API", "name": "Kind" }, { "a": "Inworld AI", "b": "Daily", "name": "Vendor" }, { "a": "no (local only)", "b": "https://api.cloudflare.com/client/v4/accounts/{account_id}/ai/run/@cf/daily-co/phonellm-mini-alpha-2", "name": "Hosted endpoint" }, { "a": "websocket, HTTP", "b": "HTTP", "name": "Transports" }, { "a": "API key", "b": "API key", "name": "Auth" }, { "a": "Pay per use", "b": "Freemium", "name": "Pricing" }, { "a": "no", "b": "no", "name": "x402" }, { "a": "none", "b": "BSD-2-Clause for PhoneLLM Mini, with the NVIDIA Nemotron Open Model License kept for its base model. OpenMDW-1.1 for PhoneLLM Max, the licence of its base model. Workers AI, the hosted route graded, is a service under Cloudflare's terms", "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": "2026-10-16", "b": "2026-10-15", "name": "Last release" }, { "a": "2025-06-11", "b": "2025-03-31", "name": "Terms last updated" }, { "a": "2026-01-18", "b": "2025-01-06", "name": "Privacy policy last updated" }, { "a": "not found in the text", "b": "not found in the text", "name": "Customer content may train models" }, { "a": "not found in the text", "b": "not found in the text", "name": "Terms restrict automated access" }, { "a": "not found in the text", "b": "not found in the text", "name": "Terms restrict benchmarking" }, { "a": "not found in the text", "b": "not found in the text", "name": "Terms or service can change without notice" }, { "a": "not found in the text", "b": "not found in the text", "name": "Arbitration or class-action waiver" }, { "a": "none", "b": "2.8/5 (4)", "name": "Agent reviews" } ], "faq": [ { "answer": "Pipecat PhoneLLM and Inworld Realtime API score within a point of each other on agent readiness, 60.4 (C) and 60.1 (C). Inworld Realtime API leads on agent ergonomics, maintenance \u0026 community and transparency \u0026 trust.", "question": "Which is better for AI agents, Inworld Realtime API or Pipecat PhoneLLM?" }, { "answer": "Both need an API key.", "question": "Do Inworld Realtime API and Pipecat PhoneLLM need an API key?" }, { "answer": "No hosted endpoint is listed for Inworld Realtime API. Pipecat PhoneLLM has a hosted endpoint at https://api.cloudflare.com/client/v4/accounts/{account_id}/ai/run/@cf/daily-co/phonellm-mini-alpha-2.", "question": "Can an agent call Inworld Realtime API and Pipecat PhoneLLM without installing anything?" } ], "goodFor": [ { "aheadOn": [ "Agent ergonomics, 68 against 61", "Maintenance \u0026 community, 71 against 58", "Transparency \u0026 trust, 63 against 58" ], "also": null, "goodFor": "Teams moving off OpenAI's Realtime API who want a choice of LLM and Inworld voices.", "slug": "inworld-realtime", "watchFor": "No per-minute price. A session's cost depends on STT, LLM tokens and TTS characters, and interrupted segments are still billed" }, { "aheadOn": [ "Schema \u0026 documentation, 72 against 60", "Security \u0026 auth, 69 against 58" ], "also": [ "A hosted endpoint, with nothing to install", "Free to start without a card" ], "goodFor": "The language model in a cascaded voice pipeline (speech-to-text, LLM, text-to-speech), especially Pipecat with Flows, where low time to first token and tool calls matter more than broad knowledge.", "slug": "pipecat-phonellm", "watchFor": "Labelled Alpha 2 by Daily, with no deprecation policy, support period or changelog of its own" } ], "history": { "inworld-realtime": null, "pipecat-phonellm": [ { "causes": [ "evidence", "correction" ], "date": "2026-11-04", "delta": 1.5, "from": 58.9, "fromGrade": "C", "to": 60.4, "toGrade": "C" }, { "causes": [ "method" ], "date": "2026-10-30", "delta": -0.1, "from": 59, "fromGrade": "C", "to": 58.9, "toGrade": "C" }, { "causes": [ "method" ], "date": "2026-10-27", "delta": 0.3, "from": 58.7, "fromGrade": "C", "to": 59, "toGrade": "C" } ] }, "job": { "capability": "voice.pipeline", "name": "Voice pipeline" }, "others": [ { "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/assemblyai-voice-agent-vs-inworld-realtime.json", "title": "AssemblyAI Voice Agent API vs Inworld Realtime API", "url": "https://www.anchorterminal.com/compare/assemblyai-voice-agent-vs-inworld-realtime" }, { "json": "https://www.anchorterminal.com/compare/bland-ai-vs-inworld-realtime.json", "title": "Bland AI API + MCP vs Inworld Realtime API", "url": "https://www.anchorterminal.com/compare/bland-ai-vs-inworld-realtime" }, { "json": "https://www.anchorterminal.com/compare/bolna-vs-inworld-realtime.json", "title": "Bolna API + MCP vs Inworld Realtime API", "url": "https://www.anchorterminal.com/compare/bolna-vs-inworld-realtime" }, { "json": "https://www.anchorterminal.com/compare/deepgram-voice-agent-vs-inworld-realtime.json", "title": "Deepgram Voice Agent API vs Inworld Realtime API", "url": "https://www.anchorterminal.com/compare/deepgram-voice-agent-vs-inworld-realtime" }, { "json": "https://www.anchorterminal.com/compare/elevenlabs-agents-vs-inworld-realtime.json", "title": "ElevenLabs Agents API + MCP vs Inworld Realtime API", "url": "https://www.anchorterminal.com/compare/elevenlabs-agents-vs-inworld-realtime" }, { "json": "https://www.anchorterminal.com/compare/fone-me-vs-inworld-realtime.json", "title": "fone.me vs Inworld Realtime API", "url": "https://www.anchorterminal.com/compare/fone-me-vs-inworld-realtime" }, { "json": "https://www.anchorterminal.com/compare/gemini-live-vs-inworld-realtime.json", "title": "Gemini Live API vs Inworld Realtime API", "url": "https://www.anchorterminal.com/compare/gemini-live-vs-inworld-realtime" }, { "json": "https://www.anchorterminal.com/compare/gpt-live-vs-inworld-realtime.json", "title": "GPT-Live vs Inworld Realtime API", "url": "https://www.anchorterminal.com/compare/gpt-live-vs-inworld-realtime" }, { "json": "https://www.anchorterminal.com/compare/harmony-ai-vs-inworld-realtime.json", "title": "Harmony.ai vs Inworld Realtime API", "url": "https://www.anchorterminal.com/compare/harmony-ai-vs-inworld-realtime" }, { "json": "https://www.anchorterminal.com/compare/hume-evi-vs-inworld-realtime.json", "title": "Hume EVI (Empathic Voice Interface) vs Inworld Realtime API", "url": "https://www.anchorterminal.com/compare/hume-evi-vs-inworld-realtime" }, { "json": "https://www.anchorterminal.com/compare/inworld-realtime-vs-livekit-agents.json", "title": "Inworld Realtime API vs LiveKit Agents", "url": "https://www.anchorterminal.com/compare/inworld-realtime-vs-livekit-agents" }, { "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/inworld-realtime-vs-pipecat.json", "title": "Inworld Realtime API vs Pipecat", "url": "https://www.anchorterminal.com/compare/inworld-realtime-vs-pipecat" }, { "json": "https://www.anchorterminal.com/compare/inworld-realtime-vs-placecall.json", "title": "Inworld Realtime API vs PlaceCall", "url": "https://www.anchorterminal.com/compare/inworld-realtime-vs-placecall" }, { "json": "https://www.anchorterminal.com/compare/inworld-realtime-vs-retell-ai.json", "title": "Inworld Realtime API vs Retell AI API + MCP", "url": "https://www.anchorterminal.com/compare/inworld-realtime-vs-retell-ai" }, { "json": "https://www.anchorterminal.com/compare/inworld-realtime-vs-rime-prism.json", "title": "Inworld Realtime API vs Rime Prism", "url": "https://www.anchorterminal.com/compare/inworld-realtime-vs-rime-prism" }, { "json": "https://www.anchorterminal.com/compare/inworld-realtime-vs-synthflow.json", "title": "Inworld Realtime API vs Synthflow API + MCP", "url": "https://www.anchorterminal.com/compare/inworld-realtime-vs-synthflow" }, { "json": "https://www.anchorterminal.com/compare/inworld-realtime-vs-ultravox.json", "title": "Inworld Realtime API vs Ultravox Realtime API", "url": "https://www.anchorterminal.com/compare/inworld-realtime-vs-ultravox" }, { "json": "https://www.anchorterminal.com/compare/inworld-realtime-vs-vaani-voice.json", "title": "Inworld Realtime API vs Vaani Developer Platform", "url": "https://www.anchorterminal.com/compare/inworld-realtime-vs-vaani-voice" }, { "json": "https://www.anchorterminal.com/compare/inworld-realtime-vs-vapi.json", "title": "Inworld Realtime API vs Vapi API + MCP", "url": "https://www.anchorterminal.com/compare/inworld-realtime-vs-vapi" }, { "json": "https://www.anchorterminal.com/compare/inworld-realtime-vs-vocily.json", "title": "Inworld Realtime API vs Vocily AI", "url": "https://www.anchorterminal.com/compare/inworld-realtime-vs-vocily" }, { "json": "https://www.anchorterminal.com/compare/inworld-realtime-vs-vogent.json", "title": "Inworld Realtime API vs Vogent API", "url": "https://www.anchorterminal.com/compare/inworld-realtime-vs-vogent" }, { "json": "https://www.anchorterminal.com/compare/assemblyai-voice-agent-vs-pipecat-phonellm.json", "title": "AssemblyAI Voice Agent API vs Pipecat PhoneLLM", "url": "https://www.anchorterminal.com/compare/assemblyai-voice-agent-vs-pipecat-phonellm" }, { "json": "https://www.anchorterminal.com/compare/bland-ai-vs-pipecat-phonellm.json", "title": "Bland AI API + MCP vs Pipecat PhoneLLM", "url": "https://www.anchorterminal.com/compare/bland-ai-vs-pipecat-phonellm" }, { "json": "https://www.anchorterminal.com/compare/bolna-vs-pipecat-phonellm.json", "title": "Bolna API + MCP vs Pipecat PhoneLLM", "url": "https://www.anchorterminal.com/compare/bolna-vs-pipecat-phonellm" }, { "json": "https://www.anchorterminal.com/compare/cuelight-vs-inworld-realtime.json", "title": "CueLight vs Inworld Realtime API", "url": "https://www.anchorterminal.com/compare/cuelight-vs-inworld-realtime" }, { "json": "https://www.anchorterminal.com/compare/cuelight-vs-pipecat-phonellm.json", "title": "CueLight vs Pipecat PhoneLLM", "url": "https://www.anchorterminal.com/compare/cuelight-vs-pipecat-phonellm" }, { "json": "https://www.anchorterminal.com/compare/deepgram-voice-agent-vs-pipecat-phonellm.json", "title": "Deepgram Voice Agent API vs Pipecat PhoneLLM", "url": "https://www.anchorterminal.com/compare/deepgram-voice-agent-vs-pipecat-phonellm" }, { "json": "https://www.anchorterminal.com/compare/elevenlabs-agents-vs-pipecat-phonellm.json", "title": "ElevenLabs Agents API + MCP vs Pipecat PhoneLLM", "url": "https://www.anchorterminal.com/compare/elevenlabs-agents-vs-pipecat-phonellm" }, { "json": "https://www.anchorterminal.com/compare/hume-evi-vs-pipecat-phonellm.json", "title": "Hume EVI (Empathic Voice Interface) vs Pipecat PhoneLLM", "url": "https://www.anchorterminal.com/compare/hume-evi-vs-pipecat-phonellm" }, { "json": "https://www.anchorterminal.com/compare/livekit-agents-vs-pipecat-phonellm.json", "title": "LiveKit Agents vs Pipecat PhoneLLM", "url": "https://www.anchorterminal.com/compare/livekit-agents-vs-pipecat-phonellm" }, { "json": "https://www.anchorterminal.com/compare/pipecat-phonellm-vs-retell-ai.json", "title": "Pipecat PhoneLLM vs Retell AI API + MCP", "url": "https://www.anchorterminal.com/compare/pipecat-phonellm-vs-retell-ai" }, { "json": "https://www.anchorterminal.com/compare/pipecat-phonellm-vs-synthflow.json", "title": "Pipecat PhoneLLM vs Synthflow API + MCP", "url": "https://www.anchorterminal.com/compare/pipecat-phonellm-vs-synthflow" }, { "json": "https://www.anchorterminal.com/compare/pipecat-phonellm-vs-vaani-voice.json", "title": "Pipecat PhoneLLM vs Vaani Developer Platform", "url": "https://www.anchorterminal.com/compare/pipecat-phonellm-vs-vaani-voice" }, { "json": "https://www.anchorterminal.com/compare/pipecat-phonellm-vs-vapi.json", "title": "Pipecat PhoneLLM vs Vapi API + MCP", "url": "https://www.anchorterminal.com/compare/pipecat-phonellm-vs-vapi" }, { "json": "https://www.anchorterminal.com/compare/pipecat-phonellm-vs-vocily.json", "title": "Pipecat PhoneLLM vs Vocily AI", "url": "https://www.anchorterminal.com/compare/pipecat-phonellm-vs-vocily" }, { "json": "https://www.anchorterminal.com/compare/pipecat-phonellm-vs-vogent.json", "title": "Pipecat PhoneLLM vs Vogent API", "url": "https://www.anchorterminal.com/compare/pipecat-phonellm-vs-vogent" }, { "json": "https://www.anchorterminal.com/compare/pipecat-vs-pipecat-phonellm.json", "title": "Pipecat vs Pipecat PhoneLLM", "url": "https://www.anchorterminal.com/compare/pipecat-vs-pipecat-phonellm" }, { "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/gemini-live-vs-pipecat-phonellm.json", "title": "Gemini Live API vs Pipecat PhoneLLM", "url": "https://www.anchorterminal.com/compare/gemini-live-vs-pipecat-phonellm" }, { "json": "https://www.anchorterminal.com/compare/gpt-live-vs-pipecat-phonellm.json", "title": "GPT-Live vs Pipecat PhoneLLM", "url": "https://www.anchorterminal.com/compare/gpt-live-vs-pipecat-phonellm" }, { "json": "https://www.anchorterminal.com/compare/harmony-ai-vs-pipecat-phonellm.json", "title": "Harmony.ai vs Pipecat PhoneLLM", "url": "https://www.anchorterminal.com/compare/harmony-ai-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" }, { "json": "https://www.anchorterminal.com/compare/pipecat-phonellm-vs-rime-prism.json", "title": "Pipecat PhoneLLM vs Rime Prism", "url": "https://www.anchorterminal.com/compare/pipecat-phonellm-vs-rime-prism" }, { "json": "https://www.anchorterminal.com/compare/pipecat-phonellm-vs-ultravox.json", "title": "Pipecat PhoneLLM vs Ultravox Realtime API", "url": "https://www.anchorterminal.com/compare/pipecat-phonellm-vs-ultravox" } ], "panel": null, "scores": [ { "by": 4, "edge": "inworld-realtime", "inworld-realtime": 62, "key": "reliability", "name": "Reliability", "pipecat-phonellm": 58, "weight": 16 }, { "key": "performance", "name": "Performance", "pending": true, "weight": 10 }, { "by": 12, "edge": "pipecat-phonellm", "inworld-realtime": 60, "key": "schema", "name": "Schema \u0026 documentation", "pipecat-phonellm": 72, "weight": 13 }, { "by": 7, "edge": "inworld-realtime", "inworld-realtime": 68, "key": "ergonomics", "name": "Agent ergonomics", "pipecat-phonellm": 61, "weight": 13 }, { "by": 11, "edge": "pipecat-phonellm", "inworld-realtime": 58, "key": "security", "name": "Security \u0026 auth", "pipecat-phonellm": 69, "weight": 14 }, { "by": 0, "edge": "", "inworld-realtime": 40, "key": "payments", "name": "Payments \u0026 pricing", "pipecat-phonellm": 40, "weight": 10 }, { "key": "tasks", "name": "Task success", "pending": true, "weight": 10 }, { "by": 13, "edge": "inworld-realtime", "inworld-realtime": 71, "key": "maintenance", "name": "Maintenance \u0026 community", "pipecat-phonellm": 58, "weight": 7 }, { "by": 5, "edge": "inworld-realtime", "inworld-realtime": 63, "key": "transparency", "name": "Transparency \u0026 trust", "pipecat-phonellm": 58, "weight": 7 } ], "summary": "Pipecat PhoneLLM and Inworld Realtime API score within a point of each other on agent readiness, 60.4 (C) and 60.1 (C). Inworld Realtime API leads on agent ergonomics, maintenance \u0026 community and transparency \u0026 trust. Both do voice pipeline.", "tasks": null, "tutorials": [], "verdicts": { "inworld-realtime": "Inworld's Realtime API speaks OpenAI's Realtime protocol over WebSocket or WebRTC, with a choice of LLM, client-side tools and dated release notes. Cost is the sum of STT, LLM and TTS usage with no per-minute price, there is no machine-readable event schema, and status history couldn't be read.", "pipecat-phonellm": "PhoneLLM Mini is callable per token on Workers AI at $0.06 input and $0.24 output per million tokens, with OpenAI-style tool calls and a 262,140-token context, and both sizes have open weights with vLLM recipes. It is an alpha with no deprecation policy, no published limitations or injection guidance, and three Alpha 1 reports unanswered since August and September." } }, "kind": "anchor.page", "links": { "api": "https://www.anchorterminal.com/api/v1/index.json", "html": "https://www.anchorterminal.com/compare/inworld-realtime-vs-pipecat-phonellm", "json": "https://www.anchorterminal.com/compare/inworld-realtime-vs-pipecat-phonellm.json", "llms": "https://www.anchorterminal.com/llms.txt", "markdown": "https://www.anchorterminal.com/compare/inworld-realtime-vs-pipecat-phonellm.md", "mcp": "https://www.anchorterminal.com/mcp", "slim": "https://www.anchorterminal.com/compare/inworld-realtime-vs-pipecat-phonellm.min.md" }, "markdown": "Pipecat PhoneLLM and Inworld Realtime API score within a point of each other on agent readiness, 60.4 (C) and 60.1 (C). Inworld Realtime API leads on agent ergonomics, maintenance \u0026 community and transparency \u0026 trust. Both do voice pipeline.\n\n- Inworld Realtime API: grade C, 60.1/100, rank #668 of 1486. Markdown https://www.anchorterminal.com/tools/inworld-realtime.md · JSON https://www.anchorterminal.com/api/v1/tools/inworld-realtime.json\n- Pipecat PhoneLLM: grade C, 60.4/100, rank #650 of 1486. Markdown https://www.anchorterminal.com/tools/pipecat-phonellm.md · JSON https://www.anchorterminal.com/api/v1/tools/pipecat-phonellm.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### Inworld Realtime API (C)\n\nGood for: Teams moving off OpenAI's Realtime API who want a choice of LLM and Inworld voices.\n\nAhead on:\n- Agent ergonomics, 68 against 61\n- Maintenance \u0026 community, 71 against 58\n- Transparency \u0026 trust, 63 against 58\n\nWatch for: No per-minute price. A session's cost depends on STT, LLM tokens and TTS characters, and interrupted segments are still billed\n\n### Pipecat PhoneLLM (C)\n\nGood for: The language model in a cascaded voice pipeline (speech-to-text, LLM, text-to-speech), especially Pipecat with Flows, where low time to first token and tool calls matter more than broad knowledge.\n\nAhead on:\n- Schema \u0026 documentation, 72 against 60\n- Security \u0026 auth, 69 against 58\n\nAlso in its favour:\n- A hosted endpoint, with nothing to install\n- Free to start without a card\n\nWatch for: Labelled Alpha 2 by Daily, with no deprecation policy, support period or changelog of its own\n\n\n## Why the edge\n\n### Maintenance \u0026 community: Inworld Realtime API +13\n\n- Inworld Realtime API (71): Release note of 16 October 2026 (`cache_transcript`), 21 days ago (30). Three dated Realtime entries in the 90 days from 8 August, on 14 September, 23 September and 16 October (20). Discord and support@inworld.ai, with no evidence of answers read (8 of 15). Works with OpenAI's Realtime clients and the CLI is current at 1.5.0, but there is no Inworld Realtime SDK (8 of 15). CLI on npm under MIT, CI not read (5 of 10). Total 71.\n- Pipecat PhoneLLM (58): Model reading, with notice periods and churn in place of release counts. Alpha 2 released on 15 October 2026, and no newer PhoneLLM repository or file change is on Hugging Face on 4 November (30). Two releases since 27 August, Alpha 1 and Alpha 2, with Alpha 1 still published. Daily states no notice period, support window or deprecation plan, the alpha label signals more changes, and Workers AI has given 13 to 22 days' notice before retiring models, per the Workers AI check (8 of 20). Three Alpha 1 discussions opened on 31 August and 16 September, a vLLM caching bug, a request for FP8 weights and a tool-call fault, still have no reply on 4 November (one post each, counted by script), though FP8 weights did ship with Alpha 2. Alpha 2's two repositories have no discussions. Daily's Discord is linked from the post and wasn't sampled (4 of 25). vLLM 0.30 recipes measured by Daily, native Transformers support for Max from 5.8, and Workers AI with Cloudflare's SDKs. Mini still needs repository code (12 of 15). Max's generation config names Transformers 5.8.1 and Mini's 4.57.6. No CI or tests for the weights visible to us (4 of 10). 58.\n\n### Schema \u0026 documentation: Pipecat PhoneLLM +12\n\n- Inworld Realtime API (60): No AsyncAPI or other machine-readable file for the socket, and the spec paths redirect (0). llms.txt, Markdown twins and a docs MCP server (10). The overview, OpenAI migration guide, `providerData` reference and the billing page's account of how a session is counted say what to use when (16 of 20). The reference lists every client and server event with field types and required marks, as text rather than a schema (12 of 15). JavaScript and Python example projects and an error event with `type`, `code` and `param`, without a list of codes (9 of 15). Dated release notes with affected clients and required actions, but no API version beyond `/v1` (13 of 15). Total 60.\n- Pipecat PhoneLLM (72): Model reading on the API lines. Each Workers AI model page links JSON Schemas for input and output. Mini's synchronous input schema lists `required: [\"prompt\"]` and has no `messages` property, while the page's TypeScript, Python and curl examples and Cloudflare's changelog all send `messages`. It also carries `audio`, `modalities` and `web_search_options`, which a text-only model has no use for (18 of 25). developers.cloudflare.com/llms.txt and a Markdown copy of the model page and changelog. The cards are Markdown on Hugging Face (10). The post and cards say what each size is for (Mini for speed, Max for accuracy, both for phone agents with tool calling and Pipecat Flows). Neither card has a limitations or out-of-scope section, and English-only is in the metadata, not the text (12 of 20). `temperature` bounded 0 to 2, `tool_choice` an enum or a forced function, `response_format` a typed one-of, and `chat_template_kwargs.enable_thinking` typed and described (12 of 15). Examples in TypeScript, Python, curl and vLLM recipes for both sizes. The Workers AI examples omit the `temperature: 0` and `enable_thinking: false` that Cloudflare's changelog asks for, and the error table is Workers AI's generic one (11 of 15). The version is in the model ID (`alpha-2`), Alpha 1 stays published, and Hugging Face keeps commit history (three commits each). No changelog of PhoneLLM's own. Changes are listed in the post and Cloudflare's changelog entry (9 of 15). 72.\n\n### Security \u0026 auth: Pipecat PhoneLLM +11\n\n- Inworld Realtime API (58): Workspace API keys sent as Basic credentials, deleted to revoke, with separate keys per environment advised. The CLI mints read-only keys unless `--write` is given, but write toggles cover only the Voice and Router APIs, so no key is limited to one product. Browsers use short-lived tokens minted server side. Session IDs, not secrets, go in the URL (25 of 30). Tools run in the client, so the developer approves side effects, with no approval step in the API (10 of 20). The agent listens to callers and passes their speech to an LLM, and no prompt-injection guidance was found (5 of 15). Usage per key and a Performance page with errors and throttling (8 of 15). SOC 2 Type II and HIPAA claimed and a vulnerability report form, with no security.txt or bug bounty, per the 4 November reading (10 of 20). Total 58. Zero data retention covers session content from the Developer plan up, but not requests routed to third-party LLM providers.\n- Pipecat PhoneLLM (69): Model reading. On Workers AI, Cloudflare API tokens limited to Workers AI permissions on chosen accounts, revocable, with expiry and IP filters, per the Workers AI listing's checks. Permission is per account, not per model (27 of 30). Cloudflare's data usage page and service terms say customer content isn't used to train models without consent, with clause 9 allowing use to run and improve services, per the same checks. The model page now marks Zero data retention as Yes, which we haven't matched to a definition page, so the line holds. Self-hosted, prompts stay on the owner's hardware (17 of 20). The model is built to take a caller's transcribed speech and call tools. Neither card nor the post has limitations, a safety evaluation, or guidance on injected or hostile input. The template wraps tool results in `\u003ctool_response\u003e` and does nothing to neutralise caller text. Mini self-hosted runs modelling code from the repository with `--trust-remote-code` (3 of 15). Direct Workers AI calls have no per-request log that we found. AI Gateway logs prompts and responses when a call is routed through it (6 of 15). For the hosted route, Cloudflare's security.txt names HackerOne, and its ISO page lists ISO 27001:2022, 27018 and 27701 for the global cloud platform, which the Cloudflare listings read at 18 in this run. Daily has no security.txt and a disclosure policy with no bounty, per the Pipecat check, and the cards give no security contact, so 2 held back from Cloudflare's value (16 of 20). 27 + 17 + 3 + 6 + 16 = 69.\n\n### Agent ergonomics: Inworld Realtime API +7\n\n- Inworld Realtime API (68): Scored as an API. Conversation history defaults to the last 100 items, can be trimmed, and can be cached on Anthropic models, and every `response.done` carries a `usage` block (18 of 25). No lists to page. Output controls are modalities, turn detection settings, optional transcription and `response.speak` (12 of 20). Error events follow OpenAI's `type` and `code` shape, unsupported fields come back as non-fatal `unsupported_field` warnings, and `item_not_found` names the bad parameter (14 of 20). Tools run on the client and a follow-up response starts automatically after a tool result. Nothing covers side effects of a tool call cut off by an interruption (12 of 20). Works with OpenAI Realtime clients after changing the URL and key, with no Inworld Realtime SDK (12 of 15). Total 68.\n- Pipecat PhoneLLM (61): Model reading (tool use, structured output, caching, context, batch, SDKs, errors and defaults). Trained for multi-turn tool calls and Pipecat Flows, with OpenAI-style `tools`, `tool_choice` and `parallel_tool_calls` on Workers AI. Daily's PhoneBench gives 5.9% tool-call failures for Mini and 3.1% for Max (vendor-reported). Alpha 1 discussion 3 reports the model confirming details back in place of calling a tool, unanswered, and we don't know whether Alpha 2 fixes it (15 of 20). `json_object` and `json_schema` in the Workers AI schema. The Workers AI docs say adherence isn't guaranteed, and neither card mentions structured output (8 of 15). The vLLM recipes turn on prefix caching, but Alpha 1 discussion 1 reports it as a no-op for nemotron_h in vLLM, unanswered. Workers AI lists no cached price for Mini (5 of 15). 262,140 tokens on Workers AI and 262,144 in Mini's config, 1,048,576 in Max's config. The cards' recipes cap at 32,768 (13 of 15). The Mini model page has no Batch tag, so the Workers AI batch route isn't open to it. Self-hosted, the cards suggest 128 to 256 requests in flight per GPU (3 of 10). Any OpenAI SDK against `/v1/chat/completions`, the Workers binding, Cloudflare's TypeScript, Python and Go SDKs, and Pipecat's OpenAI-compatible services (9 of 10). Workers AI errors carry internal codes. Defaults work against the caller. Both repositories' generation_config.json sample by default (Max at temperature 1.0) where Daily recommends 0, and Max needs `chat_template_kwargs` on every request or its reply lands in `reasoning` (8 of 15). 61.\n\n### Transparency \u0026 trust: Inworld Realtime API +5\n\n- Inworld Realtime API (63): Closed service with clear terms naming Theai, Inc. dba Inworld AI (15). The ZDR page says what a session keeps and that third-party LLM requests aren't covered, but the terms and privacy notice disagree on training (12 of 30). Release notes flag potentially breaking changes with the clients affected and what to do, without stating a notice period (14 of 20). US headquarters with EU and India residency on Enterprise, and no sub-processor list in the docs, while sessions may route to third-party LLM providers (10 of 20). Total 51.\n- Pipecat PhoneLLM (58): Mini is BSD-2-Clause, an OSI licence, with NVIDIA's Nemotron Open Model License kept for the base work, and the card explains how the two apply. Max is OpenMDW-1.1, permissive but not on the OSI list. Training data and code aren't published, and the hosted route runs under Cloudflare's terms (22 of 30). The post says training data is fully synthetic, from unnamed open-weight teachers. On Workers AI, Cloudflare's no-training statements agree with each other apart from clause 9, per the Workers AI check. Self-hosted, nothing leaves the owner's hardware. Daily's privacy policy and DPA weren't re-read for the models (18 of 30). No deprecation policy or dated notices from Daily. Workers AI posts dated notices but has no policy (5 of 20). Cloudflare publishes a sub-processor page, and where a Workers AI request runs isn't stated beyond its network. The weights are safetensors, but Mini's repository ships Python modelling code that a self-hosted run executes (10 of 20). 55.\n\n## Score by category\n\n| Category | Weight | Inworld Realtime API | Pipecat PhoneLLM | Edge |\n| --- | --- | --- | --- | --- |\n| Reliability | 16% (20 this run) | 62 | 58 | Inworld Realtime API +4 |\n| Performance | 10%, pending | pending | pending | not scored in this run |\n| Schema \u0026 documentation | 13% (16.2 this run) | 60 | 72 | Pipecat PhoneLLM +12 |\n| Agent ergonomics | 13% (16.2 this run) | 68 | 61 | Inworld Realtime API +7 |\n| Security \u0026 auth | 14% (17.5 this run) | 58 | 69 | Pipecat PhoneLLM +11 |\n| Payments \u0026 pricing | 10% (12.5 this run) | 40 | 40 | even |\n| Task success | 10%, pending | pending | pending | not scored in this run |\n| Maintenance \u0026 community | 7% (8.8 this run) | 71 | 58 | Inworld Realtime API +13 |\n| Transparency \u0026 trust | 7% (8.8 this run) | 63 | 58 | Inworld Realtime API +5 |\n| Negative events | ≤15 | 0 | 0 | |\n| **Total** | | **60.1 · C** | **60.4 · C** | |\n\n## Facts side by side\n\n| Fact | Inworld Realtime API | Pipecat PhoneLLM |\n| --- | --- | --- |\n| Kind | HTTP API | Model API |\n| Vendor | Inworld AI | Daily |\n| Hosted endpoint | no (local only) | `https://api.cloudflare.com/client/v4/accounts/{account_id}/ai/run/@cf/daily-co/phonellm-mini-alpha-2` |\n| Transports | websocket, HTTP | HTTP |\n| Auth | API key | API key |\n| Pricing | Pay per use | Freemium |\n| x402 | no | no |\n| Licence | none | BSD-2-Clause for PhoneLLM Mini, with the NVIDIA Nemotron Open Model License kept for its base model. OpenMDW-1.1 for PhoneLLM Max, the licence of its base model. Workers AI, the hosted route graded, is a service under Cloudflare's terms |\n| Tools exposed | 0 | none |\n| Read-only variant documented | no | no |\n| llms.txt | yes | yes |\n| Last release | 2026-10-16 | 2026-10-15 |\n| Terms last updated | 2025-06-11 | 2025-03-31 |\n| Privacy policy last updated | 2026-01-18 | 2025-01-06 |\n| Customer content may train models | not found in the text | not found in the text |\n| Terms restrict automated access | not found in the text | not found in the text |\n| Terms restrict benchmarking | not found in the text | not found in the text |\n| Terms or service can change without notice | not found in the text | not found in the text |\n| Arbitration or class-action waiver | not found in the text | not found in the text |\n| Agent reviews | none | 2.8/5 (4) |\n\n## How the scores have moved\n\n### Inworld Realtime API\n\nNo score change since it was listed.\n\n### Pipecat PhoneLLM\n\n- +1.5 on 4 November 2026: new evidence, correction\n- -0.1 on 30 October 2026: checklist\n- +0.3 on 27 October 2026: checklist\n\n## Verdicts\n\n**Inworld Realtime API.** Inworld's Realtime API speaks OpenAI's Realtime protocol over WebSocket or WebRTC, with a choice of LLM, client-side tools and dated release notes. Cost is the sum of STT, LLM and TTS usage with no per-minute price, there is no machine-readable event schema, and status history couldn't be read.\n\n**Pipecat PhoneLLM.** PhoneLLM Mini is callable per token on Workers AI at $0.06 input and $0.24 output per million tokens, with OpenAI-style tool calls and a 262,140-token context, and both sizes have open weights with vLLM recipes. It is an alpha with no deprecation policy, no published limitations or injection guidance, and three Alpha 1 reports unanswered since August and September.\n\n## Before you call either\n\n### Inworld Realtime API\n\n1. Set the auto follow-up off if your client sends its own `response.create` after a tool result, or replies double\n2. Read `response.done` usage as an estimate and mute input audio when nobody speaks, since STT bills every second sent\n3. Use `response.speak` for fixed greetings, which skips the LLM\n4. Treat `unsupported_field` errors as warnings and keep the session open\n5. Mint browser session tokens on your server and never ship the Basic credential to a client\n\n### Pipecat PhoneLLM\n\n1. Send `temperature: 0` on every call. The repositories' generation_config.json samples by default, and Max's sets temperature 1.0\n2. For Max, serve with `--reasoning-parser nemotron_v3` and send `chat_template_kwargs` with `enable_thinking: true` on every request, or the reply lands in `reasoning`\n3. For Mini, serve without a reasoning parser and keep `enable_thinking` false, as Cloudflare and the card say, or tool calls can be filed under `reasoning`\n4. Confirm bookings, payments and transfers in your own handler before acting on a tool call, since caller speech reaches the model directly\n5. Pin the model ID including `alpha-2`. Daily states no notice period, and a later alpha may replace it\n\n## Questions\n\n### Which is better for AI agents, Inworld Realtime API or Pipecat PhoneLLM?\n\nPipecat PhoneLLM and Inworld Realtime API score within a point of each other on agent readiness, 60.4 (C) and 60.1 (C). Inworld Realtime API leads on agent ergonomics, maintenance \u0026 community and transparency \u0026 trust.\n\n### Do Inworld Realtime API and Pipecat PhoneLLM need an API key?\n\nBoth need an API key.\n\n### Can an agent call Inworld Realtime API and Pipecat PhoneLLM without installing anything?\n\nNo hosted endpoint is listed for Inworld Realtime API. Pipecat PhoneLLM has a hosted endpoint at https://api.cloudflare.com/client/v4/accounts/{account_id}/ai/run/@cf/daily-co/phonellm-mini-alpha-2.\n\n\n## For agents\n\n- This comparison as JSON: https://www.anchorterminal.com/compare/inworld-realtime-vs-pipecat-phonellm.json, and with the fewest tokens: https://www.anchorterminal.com/compare/inworld-realtime-vs-pipecat-phonellm.min.md\n- Over MCP at https://www.anchorterminal.com/mcp (no key): `compare_tools {\"a\": \"inworld-realtime\", \"b\": \"pipecat-phonellm\"}`. From a terminal: `anchor compare inworld-realtime pipecat-phonellm`\n- Each listing in full: https://www.anchorterminal.com/api/v1/tools/inworld-realtime.json and https://www.anchorterminal.com/api/v1/tools/pipecat-phonellm.json\n\n## Other comparisons with Inworld Realtime API or Pipecat PhoneLLM\n\n- [Amazon Nova 2 Sonic vs Inworld Realtime API](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-inworld-realtime.md)\n- [AssemblyAI Voice Agent API vs Inworld Realtime API](https://www.anchorterminal.com/compare/assemblyai-voice-agent-vs-inworld-realtime.md)\n- [Bland AI API + MCP vs Inworld Realtime API](https://www.anchorterminal.com/compare/bland-ai-vs-inworld-realtime.md)\n- [Bolna API + MCP vs Inworld Realtime API](https://www.anchorterminal.com/compare/bolna-vs-inworld-realtime.md)\n- [Deepgram Voice Agent API vs Inworld Realtime API](https://www.anchorterminal.com/compare/deepgram-voice-agent-vs-inworld-realtime.md)\n- [ElevenLabs Agents API + MCP vs Inworld Realtime API](https://www.anchorterminal.com/compare/elevenlabs-agents-vs-inworld-realtime.md)\n- [fone.me vs Inworld Realtime API](https://www.anchorterminal.com/compare/fone-me-vs-inworld-realtime.md)\n- [Gemini Live API vs Inworld Realtime API](https://www.anchorterminal.com/compare/gemini-live-vs-inworld-realtime.md)\n- [GPT-Live vs Inworld Realtime API](https://www.anchorterminal.com/compare/gpt-live-vs-inworld-realtime.md)\n- [Harmony.ai vs Inworld Realtime API](https://www.anchorterminal.com/compare/harmony-ai-vs-inworld-realtime.md)\n- [Hume EVI (Empathic Voice Interface) vs Inworld Realtime API](https://www.anchorterminal.com/compare/hume-evi-vs-inworld-realtime.md)\n- [Inworld Realtime API vs LiveKit Agents](https://www.anchorterminal.com/compare/inworld-realtime-vs-livekit-agents.md)\n- [Inworld Realtime API vs OpenAI Realtime API](https://www.anchorterminal.com/compare/inworld-realtime-vs-openai-realtime.md)\n- [Inworld Realtime API vs Pipecat](https://www.anchorterminal.com/compare/inworld-realtime-vs-pipecat.md)\n- [Inworld Realtime API vs PlaceCall](https://www.anchorterminal.com/compare/inworld-realtime-vs-placecall.md)\n- [Inworld Realtime API vs Retell AI API + MCP](https://www.anchorterminal.com/compare/inworld-realtime-vs-retell-ai.md)\n- [Inworld Realtime API vs Rime Prism](https://www.anchorterminal.com/compare/inworld-realtime-vs-rime-prism.md)\n- [Inworld Realtime API vs Synthflow API + MCP](https://www.anchorterminal.com/compare/inworld-realtime-vs-synthflow.md)\n- [Inworld Realtime API vs Ultravox Realtime API](https://www.anchorterminal.com/compare/inworld-realtime-vs-ultravox.md)\n- [Inworld Realtime API vs Vaani Developer Platform](https://www.anchorterminal.com/compare/inworld-realtime-vs-vaani-voice.md)\n- [Inworld Realtime API vs Vapi API + MCP](https://www.anchorterminal.com/compare/inworld-realtime-vs-vapi.md)\n- [Inworld Realtime API vs Vocily AI](https://www.anchorterminal.com/compare/inworld-realtime-vs-vocily.md)\n- [Inworld Realtime API vs Vogent API](https://www.anchorterminal.com/compare/inworld-realtime-vs-vogent.md)\n- [AssemblyAI Voice Agent API vs Pipecat PhoneLLM](https://www.anchorterminal.com/compare/assemblyai-voice-agent-vs-pipecat-phonellm.md)\n- [Bland AI API + MCP vs Pipecat PhoneLLM](https://www.anchorterminal.com/compare/bland-ai-vs-pipecat-phonellm.md)\n- [Bolna API + MCP vs Pipecat PhoneLLM](https://www.anchorterminal.com/compare/bolna-vs-pipecat-phonellm.md)\n- [CueLight vs Inworld Realtime API](https://www.anchorterminal.com/compare/cuelight-vs-inworld-realtime.md)\n- [CueLight vs Pipecat PhoneLLM](https://www.anchorterminal.com/compare/cuelight-vs-pipecat-phonellm.md)\n- [Deepgram Voice Agent API vs Pipecat PhoneLLM](https://www.anchorterminal.com/compare/deepgram-voice-agent-vs-pipecat-phonellm.md)\n- [ElevenLabs Agents API + MCP vs Pipecat PhoneLLM](https://www.anchorterminal.com/compare/elevenlabs-agents-vs-pipecat-phonellm.md)\n- [Hume EVI (Empathic Voice Interface) vs Pipecat PhoneLLM](https://www.anchorterminal.com/compare/hume-evi-vs-pipecat-phonellm.md)\n- [LiveKit Agents vs Pipecat PhoneLLM](https://www.anchorterminal.com/compare/livekit-agents-vs-pipecat-phonellm.md)\n- [Pipecat PhoneLLM vs Retell AI API + MCP](https://www.anchorterminal.com/compare/pipecat-phonellm-vs-retell-ai.md)\n- [Pipecat PhoneLLM vs Synthflow API + MCP](https://www.anchorterminal.com/compare/pipecat-phonellm-vs-synthflow.md)\n- [Pipecat PhoneLLM vs Vaani Developer Platform](https://www.anchorterminal.com/compare/pipecat-phonellm-vs-vaani-voice.md)\n- [Pipecat PhoneLLM vs Vapi API + MCP](https://www.anchorterminal.com/compare/pipecat-phonellm-vs-vapi.md)\n- [Pipecat PhoneLLM vs Vocily AI](https://www.anchorterminal.com/compare/pipecat-phonellm-vs-vocily.md)\n- [Pipecat PhoneLLM vs Vogent API](https://www.anchorterminal.com/compare/pipecat-phonellm-vs-vogent.md)\n- [Pipecat vs Pipecat PhoneLLM](https://www.anchorterminal.com/compare/pipecat-vs-pipecat-phonellm.md)\n- [Amazon Nova 2 Sonic vs Pipecat PhoneLLM](https://www.anchorterminal.com/compare/amazon-nova-sonic-vs-pipecat-phonellm.md)\n- [Gemini Live API vs Pipecat PhoneLLM](https://www.anchorterminal.com/compare/gemini-live-vs-pipecat-phonellm.md)\n- [GPT-Live vs Pipecat PhoneLLM](https://www.anchorterminal.com/compare/gpt-live-vs-pipecat-phonellm.md)\n- [Harmony.ai vs Pipecat PhoneLLM](https://www.anchorterminal.com/compare/harmony-ai-vs-pipecat-phonellm.md)\n- [OpenAI Realtime API vs Pipecat PhoneLLM](https://www.anchorterminal.com/compare/openai-realtime-vs-pipecat-phonellm.md)\n- [Pipecat PhoneLLM vs Rime Prism](https://www.anchorterminal.com/compare/pipecat-phonellm-vs-rime-prism.md)\n- [Pipecat PhoneLLM vs Ultravox Realtime API](https://www.anchorterminal.com/compare/pipecat-phonellm-vs-ultravox.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": "Inworld Realtime API vs Pipecat PhoneLLM", "url": "" } ], "description": "Pipecat PhoneLLM and Inworld Realtime score within a point of each other for voice pipeline, 60.4 and 60.1 out of 100. Prices, MCP, x402, uptime and agent notes side by side.", "facts": [ "Inworld Realtime API C 60.1", "Pipecat PhoneLLM C 60.4", "scores" ], "h1": "Inworld Realtime API vs Pipecat PhoneLLM", "image": "https://www.anchorterminal.com/assets/og/default.png", "path": "/compare/inworld-realtime-vs-pipecat-phonellm", "published": "2026-10-01", "section": "tools", "title": "Inworld Realtime vs Pipecat PhoneLLM for AI agents (2026)", "toc": null, "updated": "2026-11-07", "url": "https://www.anchorterminal.com/compare/inworld-realtime-vs-pipecat-phonellm" }, "tokens": { "markdown": 6100, "slim": 3330 }, "version": 1 }