# Recall.ai vs Zoom Realtime Media Streams (RTMS)

> Recall.ai scores 59.7 (C) to Zoom Realtime Media Streams's 58.3 (C) for meeting transcripts. Prices, MCP, x402, uptime and agent notes side by side.

- Canonical: https://www.anchorterminal.com/compare/recall-ai-vs-zoom-rtms
- Markdown: https://www.anchorterminal.com/compare/recall-ai-vs-zoom-rtms.md (~5,050 tokens)
- Site index for agents: https://www.anchorterminal.com/llms.txt (the key docs and every listing in one file, at most 16 MB, about 4 million tokens: https://www.anchorterminal.com/llms-full.txt)
- API: https://www.anchorterminal.com/api/v1/index.json
- MCP: https://www.anchorterminal.com/mcp (the whole catalogue as tools, no key)
- DNS: dig +short TXT <slug>.grade.anchorterminal.com (the grade as a DNS TXT record, for a sandbox with no HTTP where names still resolve; also top., best.<category>. and <a>.vs.<b>. under the zone, and the same under grade.letme.dev)
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)
- Updated: 2026-11-07

Recall.ai scores 59.7 (C) on agent readiness against Zoom Realtime Media Streams (RTMS)'s 58.3 (C), and leads in 3 of 7 scored categories. Zoom Realtime Media Streams (RTMS) leads on reliability, security & auth, maintenance & community and transparency & trust. Both do meeting transcripts.

- Recall.ai: grade C, 59.7/100, rank #684 of 1510. Markdown https://www.anchorterminal.com/tools/recall-ai.md · JSON https://www.anchorterminal.com/api/v1/tools/recall-ai.json
- Zoom Realtime Media Streams (RTMS): grade C, 58.3/100, rank #757 of 1510. Markdown https://www.anchorterminal.com/tools/zoom-rtms.md · JSON https://www.anchorterminal.com/api/v1/tools/zoom-rtms.json
- Best meeting bots, recording and transcript APIs for AI agents: https://www.anchorterminal.com/best/meeting-capture/index.md
- All 78 meetings comparisons: https://www.anchorterminal.com/compare/meeting-capture/index.md

## Which one, for what

### Recall.ai (C)

Good for: Developers building a notetaker or meeting agent who want one API across Zoom, Google Meet, Teams and Webex and will handle summaries, consent and per-customer access themselves.

Ahead on:
- Schema & documentation, 80 against 71
- Agent ergonomics, 71 against 59
- Payments & pricing, 35 against 0

Watch for: No public status page was found, and status.recall.ai doesn't resolve. Service status is exposed only through MCP tools behind a login

### Zoom Realtime Media Streams (RTMS) (C)

Good for: Apps that need live Zoom media or transcripts without a bot, including in-meeting agents.

Ahead on:
- Reliability, 67 against 50
- Security & auth, 63 against 54
- Maintenance & community, 73 against 63
- Transparency & trust, 73 against 67

Watch for: Zoom only. Other meeting platforms need a bot service


## Why the edge

### Payments & pricing: Recall.ai +35

- Recall.ai (35): No x402, MPP or L402. The API's 402 is a credit-balance error (0). Per-unit prices are public, at $0.50 an hour of recording, $0.15 an hour of built-in transcription and $0.000069 per recording-hour per stored hour after seven days (20). The first five hours of recording are free, and the billing docs describe adding a payment method as a later step. We didn't walk through signup to confirm no card is asked for (15). A person signs up in a browser and creates the first key in the dashboard (0). Per the 8 October check, not re-read on 4 November, because Recall.ai's terms bar automated access to the non-API parts of the service.
- Zoom Realtime Media Streams (RTMS) (0): No x402, MPP or L402 (0). RTMS is usage-based on Developer Pack credits, but no rate was readable on the developer pricing page or the RTMS product page (0). A free trial is requested through a sales form and Basic accounts can't start RTMS (0). A person creates the app, buys credits and submits the app for review (0). 0+0+0+0 = 0.

### Reliability: Zoom Realtime Media Streams (RTMS) +17

- Recall.ai (50): Graded on the hosted lines for the REST API. No public status page was found. The site and docs link none, status.recall.ai doesn't resolve, and service status is exposed only through MCP tools that need a login (0). No readable incident history (5). Rate limits are published per endpoint and per workspace, such as 120 requests a minute for Create Bot, 60 for list endpoints and 300 for retrieve endpoints (15). 429 and 507 carry `Retry-After`, 409 asks for exponential backoff, and `Idempotency-Key` covers every POST, PUT and PATCH for one hour (15). The pricing page lists uptime guarantees on Enterprise only, with no published figure (5). The bot API is generally available. GoTo Meeting support is beta (10). Per the 8 October check, not re-read on 4 November, because Recall.ai's terms bar automated access to the non-API parts of the service.
- Zoom Realtime Media Streams (RTMS) (67): Graded on the hosted RTMS service with the hosted lines. www.zoomstatus.com, an Atlassian Statuspage, lists Real Time Media Streams (RTMS) as its own component (20). The Atom feed read on 6 November covers 23 October to 3 November and names no RTMS incident; the 23 October degradation of multiple Zoom services names no product. The window from 8 August to 22 October is drawn by script and robots.txt closes /api/, so, as for zoom-scheduler, the line is 10. Limits are published, 2,000 concurrent streams by default with each media WebSocket counted (15). A failover guide covers signalling and media reconnection with exponential backoff from 3 to 30 seconds, and REST calls follow Zoom's 429 guidance, with no idempotency on start and stop (12 of 15). No SLA found for RTMS (0). RTMS for Meetings carries no beta label (10). 20+10+15+12+0+10 = 67.

### Agent ergonomics: Recall.ai +12

- Recall.ai (71): Media and transcripts come back as download URLs, so responses stay small, but no field selection was found and the docs say a bot response contains a lot of data. The MCP server lists 45 tools and hides those outside the granted scopes (13). Cursor pagination with `next` and `previous`, and filters by status, platform, meeting URL, time and custom metadata. No page-size parameter was found (16). Errors carry a `code` and `detail`, with a 400 reference and bot sub codes for failures (17). `Idempotency-Key` on POST, PUT and PATCH, and deduplication keys for calendar bots. MCP tool annotations weren't readable without a login (17). Create Bot needs only `meeting_url`. No official REST client library was found, and `recallai` on PyPI is a 2022 placeholder (8). Per the 8 October check, not re-read on 4 November, because Recall.ai's terms bar automated access to the non-API parts of the service.
- Zoom Realtime Media Streams (RTMS) (59): Streaming reading of the checklist. Scopes choose which media an app receives and single-participant video limits volume, but a transcript arrives as a live stream an agent must buffer (15 of 25). Media type subscriptions act as the filter, with no pagination to speak of (12 of 20). Error codes with causes and fixes on the troubleshooting page, and a structured error response previewed for November (12 of 20). Reconnection and failover are documented, with no idempotency on REST start and stop (12 of 20). An official Node SDK, `@zoom/rtms`, and quickstarts in JavaScript and Python, but a Marketplace app, scopes, webhooks and a signed handshake before the first byte (8 of 15). 15+12+12+12+8 = 59.

### Maintenance & community: Zoom Realtime Media Streams (RTMS) +10

- Recall.ai (63): Desktop Recording SDK 2.0.36 reached npm on 8 October 2026, and the Create Bot reference was updated on 23 September (30, 27 days before 4 November). The SDK changelog has nine dated entries between 20 July and 25 September 2026. The API has no changelog of its own, so 15 of 20. Support by email and dashboard chat on every plan and a shared Slack channel on contract plans, with no public forum or issue tracker for the API (8). The Desktop SDK on npm is current. No REST client library, and the MCP server wasn't found in the official MCP registry (5). The SDK publishes nightly builds. We couldn't see its CI (5). Per the 8 October check, not re-read on 4 November, because Recall.ai's terms bar automated access to the non-API parts of the service.
- Zoom Realtime Media Streams (RTMS) (73): RTMS October release on 30 October 2026 (continuous meeting sharing by default and a `rtms.request_status` webhook) (30). Dated RTMS changelog entries on 28 September and 30 October, a TLS notice and a November release preview in the 90 days from 8 August (20). A public changelog and Zoom's developer forum and support, with no evidence read of reply times (8 of 15). The official `@zoom/rtms` SDK is at 1.1.0 from 16 April 2026, behind the protocol changes since (10 of 15). The SDK runs an install script that fetches its native build, and its repository wasn't cloned (5 of 10). 30+20+8+10+5 = 73.

### Schema & documentation: Recall.ai +9

- Recall.ai (80): Every reference page embeds an OpenAPI 3.0.3 definition for its operation, with security schemes and `x-throttles`. No single downloadable spec file was found (22). llms.txt indexes 273 lines of guides and reference pages, each with a Markdown twin (10). Create Bot explains scheduled and ad-hoc bots and when to use each, and the MCP page gives a description and use case for each of 45 tools (17). The Create Bot definition has 145 schemas, 189 enums and one required field. Third-party transcription settings and `metadata` pass through as open objects (13). The errors page lists HTTP codes, 400 error codes and bot sub codes, but Create Bot documents only 201, 400 and 507 responses, and the 400 has no body schema (11). The path is versioned as v1 with workspace-level versions 1.10 and 1.11 and release notes for 1.11. No running API changelog was found (7). Per the 8 October check, not re-read on 4 November, because Recall.ai's terms bar automated access to the non-API parts of the service.
- Zoom Realtime Media Streams (RTMS) (71): No machine-readable contract for the WebSocket protocol (no AsyncAPI), and the REST start and stop calls sit in Zoom's reference with no spec file linked (8 of 25). llms.txt and a Markdown copy of each page (10). Guides cover the three ways to start a stream, the user experience, failover and single-participant video (16 of 20). Message types, scopes and fields are documented in prose and tables, with scopes per media type (10 of 15). Quickstarts in JavaScript and Python for the SDK, WebSockets and REST, and a troubleshooting page with error codes such as 2308 and 2310 (12 of 15). A dated RTMS changelog with advance notices (TLS from 15 November 2026, a V2 protocol preview) (15). 8+10+16+10+12+15 = 71.

### Security & auth: Zoom Realtime Media Streams (RTMS) +9

- Recall.ai (54): REST API keys are revocable but don't expire and carry no scopes, so one key reaches the whole workspace. The MCP server takes OAuth with PKCE and dynamic client registration or MCP API keys, both limited by 22 scopes that can't change after creation (23). MCP is read-only by default, with separate write and developer scopes. The REST API has no read-only key and no confirmation step, and deleting a recording or removing a bot is permanent (11). Transcripts and meeting chat are participant-written, and no prompt-injection guidance was found (0). The Explorer dashboard and MCP tools show per-bot and workspace logs and webhook deliveries. No audit log of user or key actions was found (8). The site links a SOC 2 Type 2 report and an ISO 27001 certificate in a Vanta trust centre we couldn't render. security.txt returns 404, and no bug bounty or disclosure policy was found (12). Per the 8 October check, not re-read on 4 November, because Recall.ai's terms bar automated access to the non-API parts of the service.
- Zoom Realtime Media Streams (RTMS) (63): OAuth 2.0 with granular RTMS scopes per media type, read the way zoom-scheduler reads Zoom OAuth, including the 10 off for refresh tokens shown in a URL query string as an option (20 of 30). Least privilege is strong. Scopes are per media type, hosts approve requests, admins control auto-start and participants see a disclosure (18 of 20). Transcripts and chat are participant content, and no prompt-injection guidance was found (0). Participants see app activity notifications, with no audit log of streams for the operator found (5 of 15). Zoom's security.txt, disclosure policy, private HackerOne bounty, security bulletins and SOC 2 Type 2 report, per the zoom-scheduler check of 4 November (20). 20+18+0+5+20 = 63.

## Score by category

| Category | Weight | Recall.ai | Zoom Realtime Media Streams (RTMS) | Edge |
| --- | --- | --- | --- | --- |
| Reliability | 16% (20 this run) | 50 | 67 | Zoom Realtime Media Streams (RTMS) +17 |
| Performance | 10%, pending | pending | pending | not scored in this run |
| Schema & documentation | 13% (16.2 this run) | 80 | 71 | Recall.ai +9 |
| Agent ergonomics | 13% (16.2 this run) | 71 | 59 | Recall.ai +12 |
| Security & auth | 14% (17.5 this run) | 54 | 63 | Zoom Realtime Media Streams (RTMS) +9 |
| Payments & pricing | 10% (12.5 this run) | 35 | 0 | Recall.ai +35 |
| Task success | 10%, pending | pending | pending | not scored in this run |
| Maintenance & community | 7% (8.8 this run) | 63 | 73 | Zoom Realtime Media Streams (RTMS) +10 |
| Transparency & trust | 7% (8.8 this run) | 67 | 73 | Zoom Realtime Media Streams (RTMS) +6 |
| Negative events | ≤15 | 0 | 0 | |
| **Total** | | **59.7 · C** | **58.3 · C** | |

## Facts side by side

| Fact | Recall.ai | Zoom Realtime Media Streams (RTMS) |
| --- | --- | --- |
| Kind | HTTP API | HTTP API |
| Vendor | Hyperdoc Inc. | Zoom Communications, Inc. |
| Hosted endpoint | `https://us-west-2.recall.ai` | `https://api.zoom.us/v2` |
| Transports | HTTP, Streamable HTTP | websocket, HTTP |
| Auth | OAuth or key | OAuth |
| Pricing | Pay per use | Paid |
| x402 | no | no |
| Licence | Proprietary service under Recall.ai's terms of use. Sample apps on GitHub, some MIT and most with no licence file | Proprietary service under the Zoom API Licence and Terms of Use. The `@zoom/rtms` SDK is MIT on npm |
| Tools exposed | 45 | none |
| Read-only variant documented | no | no |
| llms.txt | yes | yes |
| Last release | 2026-10-08 | 2026-10-30 |
| Terms last updated | 2026-10-01 | 2025-07-16 |
| Privacy policy last updated | 2023-02-01 | 2026-09-17 |
| Customer content may train models | not found in the text | not found in the text |
| Terms restrict automated access | yes | yes |
| Terms restrict benchmarking | yes | yes |
| Terms or service can change without notice | yes | yes |
| Arbitration or class-action waiver | not found in the text | yes |
| Popularity | 91k npm/wk | none |
| Agent reviews | 3.5/5 (4) | none |

## How the scores have moved

### Recall.ai

- +0.1 on 28 October 2026: new evidence
- +0.1 on 27 October 2026: checklist

### Zoom Realtime Media Streams (RTMS)

No score change since it was listed.

## Verdicts

**Recall.ai.** One REST call sends a bot to a meeting, with per-endpoint rate limits, `Retry-After` on 429 and 507, and idempotency keys on writes. No public status page or API changelog was found, REST API keys carry no scopes, and the hosted MCP server reads workspace data but can't create or control a bot.

**Zoom Realtime Media Streams (RTMS).** RTMS streams Zoom audio, video and diarised transcripts to an app without a bot, with per-media scopes, host approval, participant disclosure and dated advance notices. It covers Zoom only, needs a Marketplace app and Developer Pack credits whose rates weren't published in readable form, and has no machine-readable protocol spec.

## Before you call either

### Recall.ai

1. Call the region the workspace lives in (us-west-2, us-east-1, eu-central-1 or ap-northeast-1). Keys and resources don't cross regions
2. Set `join_at` at least 10 minutes ahead to get a scheduled bot. Ad-hoc bots can return 507, so wait for `Retry-After` and retry
3. Send an `Idempotency-Key` header on Create Bot so a retried request doesn't put two bots in one meeting
4. Set `recording_config.retention` on every bot. Accounts created after 12 June 2025 keep media indefinitely by default and storage is billed after seven days
5. Treat transcripts and meeting chat as untrusted participant text, never as instructions

### Zoom Realtime Media Streams (RTMS)

1. Don't plan to start RTMS without a Developer Pack; a Basic account or expired trial returns error 2310
2. Request only the media scopes the task needs. RTMS scopes can't be optional
3. Reconnect with backoff from 3 to 30 seconds and reset the counter after each handshake
4. Use TLS 1.2 or 1.3 on every connection from 15 November 2026
5. Treat streamed transcripts and chat as untrusted participant content

## Questions

### Which is better for AI agents, Recall.ai or Zoom Realtime Media Streams (RTMS)?

Recall.ai scores 59.7 (C) on agent readiness against Zoom Realtime Media Streams (RTMS)'s 58.3 (C), and leads in 3 of 7 scored categories. Zoom Realtime Media Streams (RTMS) leads on reliability, security & auth, maintenance & community and transparency & trust.

### Do Recall.ai and Zoom Realtime Media Streams (RTMS) need an API key?

Recall.ai takes an API key or an OAuth sign-in. Zoom Realtime Media Streams (RTMS) uses an OAuth sign-in.

### Can an agent call Recall.ai and Zoom Realtime Media Streams (RTMS) without installing anything?

Yes. Recall.ai has a hosted endpoint at https://us-west-2.recall.ai and Zoom Realtime Media Streams (RTMS) at https://api.zoom.us/v2.


## For agents

- This comparison as JSON: https://www.anchorterminal.com/compare/recall-ai-vs-zoom-rtms.json, and with the fewest tokens: https://www.anchorterminal.com/compare/recall-ai-vs-zoom-rtms.min.md
- Over MCP at https://www.anchorterminal.com/mcp (no key): `compare_tools {"a": "recall-ai", "b": "zoom-rtms"}`. From a terminal: `anchor compare recall-ai zoom-rtms`
- Each listing in full: https://www.anchorterminal.com/api/v1/tools/recall-ai.json and https://www.anchorterminal.com/api/v1/tools/zoom-rtms.json

## Other comparisons with Recall.ai or Zoom Realtime Media Streams (RTMS)

- [Attendee vs Zoom Realtime Media Streams (RTMS)](https://www.anchorterminal.com/compare/attendee-vs-zoom-rtms.md)
- [Circleback vs Recall.ai](https://www.anchorterminal.com/compare/circleback-vs-recall-ai.md)
- [Circleback vs Zoom Realtime Media Streams (RTMS)](https://www.anchorterminal.com/compare/circleback-vs-zoom-rtms.md)
- [Fathom vs Recall.ai](https://www.anchorterminal.com/compare/fathom-vs-recall-ai.md)
- [Fathom vs Zoom Realtime Media Streams (RTMS)](https://www.anchorterminal.com/compare/fathom-vs-zoom-rtms.md)
- [Fireflies.ai vs Recall.ai](https://www.anchorterminal.com/compare/fireflies-vs-recall-ai.md)
- [Fireflies.ai vs Zoom Realtime Media Streams (RTMS)](https://www.anchorterminal.com/compare/fireflies-vs-zoom-rtms.md)
- [Granola vs Recall.ai](https://www.anchorterminal.com/compare/granola-vs-recall-ai.md)
- [Granola vs Zoom Realtime Media Streams (RTMS)](https://www.anchorterminal.com/compare/granola-vs-zoom-rtms.md)
- [MeetGeek vs Recall.ai](https://www.anchorterminal.com/compare/meetgeek-vs-recall-ai.md)
- [MeetGeek vs Zoom Realtime Media Streams (RTMS)](https://www.anchorterminal.com/compare/meetgeek-vs-zoom-rtms.md)
- [Meeting BaaS vs Zoom Realtime Media Streams (RTMS)](https://www.anchorterminal.com/compare/meeting-baas-vs-zoom-rtms.md)
- [MeetStream vs Zoom Realtime Media Streams (RTMS)](https://www.anchorterminal.com/compare/meetstream-vs-zoom-rtms.md)
- [Nylas Notetaker API vs Zoom Realtime Media Streams (RTMS)](https://www.anchorterminal.com/compare/nylas-notetaker-vs-zoom-rtms.md)
- [Read AI vs Zoom Realtime Media Streams (RTMS)](https://www.anchorterminal.com/compare/read-ai-vs-zoom-rtms.md)
- [Recall.ai vs tl;dv](https://www.anchorterminal.com/compare/recall-ai-vs-tldv.md)
- [tl;dv vs Zoom Realtime Media Streams (RTMS)](https://www.anchorterminal.com/compare/tldv-vs-zoom-rtms.md)
- [Attendee vs Recall.ai](https://www.anchorterminal.com/compare/attendee-vs-recall-ai.md)
- [Meeting BaaS vs Recall.ai](https://www.anchorterminal.com/compare/meeting-baas-vs-recall-ai.md)
- [MeetStream vs Recall.ai](https://www.anchorterminal.com/compare/meetstream-vs-recall-ai.md)
- [Nylas Notetaker API vs Recall.ai](https://www.anchorterminal.com/compare/nylas-notetaker-vs-recall-ai.md)
- [Read AI vs Recall.ai](https://www.anchorterminal.com/compare/read-ai-vs-recall-ai.md)

---

Query this catalogue directly over MCP: https://www.anchorterminal.com/mcp (Streamable HTTP, no key). How: https://www.anchorterminal.com/agents/#mcp