Head to head · WhatsApp messaging · October 2026 research run

Telnyx API + MCP vs Wati

Telnyx API + MCP scores 75.3 (BB) on agent readiness against Wati's 51.1 (D), and leads in every scored category. Both do whatsapp messaging.

Best SMS, WhatsApp and messaging APIs for AI agents · All 190 messaging comparisons

Which one, for what

Telnyx API + MCP BB

Best for Cost-sensitive US SMS and for agents that must onboard and pay without a person.

Ahead on

  • Reliability, 80 against 55
  • Schema & documentation, 93 against 79
  • Agent ergonomics, 90 against 48
  • Security & auth, 45 against 40
  • Payments & pricing, 65 against 20
  • Maintenance & community, 83 against 62
  • Transparency & trust, 72 against 52

Also in its favour

  • Agent-ready, a grade of BB or better
  • Runs on your own machine

Watch for

No per-key scopes or read-only keys, and the MCP can send or buy numbers without confirmation

Wati D

Best for Businesses whose customers are on WhatsApp and Instagram and who want an inbox, campaigns and an agent on one account.

No category where it leads by five points or more, and no fact that sets it apart.

Watch for

No idempotency keys; the docs say not to assume idempotency and to reconcile before resending

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Payments & pricing Telnyx API + MCP +45

Telnyx API + MCP 65

x402 exists on Telnyx's own API at /v2/x402/credit_account (USDC on Base), with MPP at /v2/machine-payments/account-credit (Stripe Link or Tempo USDC), but both top up account credit rather than pay per message. Telnyx's x402 skill says x402 sits behind an account feature flag (403 when it's off) and that per-payment minimum and maximum limits are configurable and stated in 422 errors; the MPP skill names no flag (25 of 40). Per-part SMS and MMS rates and carrier passthrough fees are published without a login, also as pricing.md (20). No free credit, a new account starts at zero (0). An agent can sign up through /v2/bot_challenge and /v2/bot_signup, confirm by a magic link sent to a mailbox it can read, mint a key at /v2/api_keys and fund the account by MPP or x402 with no browser (20).

Wati 20

No x402, MPP or L402 (0). Public plan prices only, plus Meta's fees by reference (10 of 20). A 7-day free trial whose card rule isn't stated (10 of 20). A person signs up and connects a WhatsApp Business account (0). 0 + 10 + 10 + 0 = 20.

Agent ergonomics Telnyx API + MCP +42

Telnyx API + MCP 90

List endpoints take page size and number, with no field selection. The hosted MCP lists 6 tools on 1 November (3 meta-tools that load endpoint schemas on demand and 3 read-only MCP App launchers), but we scored the API (15 of 25). Filters on message and number lists (20). JSON errors with a code, title and detail, and a documented code list (20). POST /v2/messages now declares an optional Idempotency-Key header in its reference, and so do the seven other send endpoints in the OpenAPI file (long code, short code, number pool, alphanumeric sender ID, group MMS, RCS and WhatsApp). Scheduled sends (POST /v2/messages/schedule) and account writes such as messaging profiles and number orders take none. That is the coverage the Telnyx Voice API listing earns 20 for, where command_id is on 38 of 40 call commands. Only successful responses are replayed, for up to 24 hours, a key still in progress returns 409 and a key reused with a different request returns 422. The Node SDK carries it from v7.27.0 (21 October 2026), where 17 October found none (20). A send needs from or a messaging profile, to and text. Official SDKs in Node, Python and more (15). 15 + 20 + 20 + 20 + 15 = 90.

Wati 48

The MCP catalogue lists 73 WATI tools plus Astra groups, some loaded on demand through list_skills and activate_skill (10 of 25). Page and page size with filters on lists, and bounded search (17 of 20). Error handling is documented per status, but bodies can be JSON, plain text or empty, so an agent can't rely on one shape (10 of 20). No idempotency keys; tag add and remove are idempotent, and the docs say to reconcile rather than retry sends (5 of 20). No official SDK; runnable examples in five languages, and region-specific hosts add setup (6 of 15). 10 + 17 + 10 + 5 + 6 = 48.

Reliability Telnyx API + MCP +25

Telnyx API + MCP 80

Read as a hosted API, as on 1, 6 and 17 October. Status page at status.telnyx.com (run on incident.io) with a readable incident feed and history (20). In the 90 days to 1 November, one incident Telnyx marked major hit the API broadly, intermittent 5XX responses across endpoints from 13:49 to 15:50 UTC on 23 September. Telnyx also marked two single-carrier messaging incidents major, MMS delays to AT&T (about 3 hours the same day) and delayed delivery of 10DLC SMS to AT&T from about 15:00 UTC on 6 October to 13:03 on 7 October, with messages accepted and other carriers unaffected. We count both as single-carrier routes, as on 6 and 17 October. Partial incidents read as minor. From 20:55 UTC on 5 August to 14:03 on 6 August a small percentage of Messaging API requests returned 503 or 504 (degraded performance, delivery unaffected), first read on 1 November. On 21 October, between about 18:17 and 19:20 UTC, a network fault caused intermittent failures across services, which Telnyx posted two days later with impact none, and we read it as partial, as for the Telnyx Voice API listing. Others include outbound messaging latency from 15 September, intermittent outbound errors and v2 API processing delays on 13 October, MMS delivery receipt queueing on 27 October and webhook delays on 29 and 30 October. The other majors since 17 October are other products, Programmable Video (critical, 16 October) and Wireless connectivity (21 October) (10). The 23 September major leaves the window after 22 December. pricing.md lists 50 SMS a second and 100 API requests a second on pay-as-you-go, and the messaging rate-limit page lists 50 SMS, 15 MMS and 1 RCS a second per account with per-sender limits, both read on 1 November (15). The rate-limiting page, re-read on 1 November, documents 429 with error 10011, Retry-After and x-ratelimit headers, bounded exponential backoff with jitter and retrying only safely repeatable operations, and the retries page says never to auto-retry a timed-out send without reconciling first (15). The SLA file at telnyx.com/ai/sla.json, re-read on 1 November, states 99.99 per cent for core voice and messaging with credits of 10, 25 and 50 per cent, without saying who qualifies (10). Messaging is generally available (10). 20 + 10 + 15 + 15 + 10 + 10 = 80.

Wati 55

Read with the hosted lines. status.wati.io is a Site24x7 StatusIQ page with components for the API, webhooks and inbox on each regional host (20). Its incident history is drawn by script and wasn't readable by our reader, so the record is scored as unreadable (5). Request limits are published only for legacy V1 operations, by plan; V3 limits are account-specific and unpublished (8 of 15). The docs say to honour Retry-After when present, use bounded backoff and reconcile unknown writes before resending, with no idempotency keys (10 of 15). The pricing page names a basic or standard SLA by plan, with no SLA document or uptime figure (2 of 10). The V3 API is the documented default (10). 20 + 5 + 8 + 10 + 2 + 10 = 55.

Maintenance & community Telnyx API + MCP +21

Telnyx API + MCP 83

telnyx-node v7.29.0 and telnyx-python v4.190.0 both tagged on 2026-10-28 (30). 17 telnyx-node releases in the 90 days to 1 November, from v7.13.0 on 4 August to v7.29.0, and 20 telnyx-python tags in the same window, with dated release notes up to 30 October 2026, messaging entries among them on 26 October (MM7 gateway for MMS) and 29 October (automatic SMS chunking) (20). Closed service with public release notes and a support channel. We couldn't read GitHub issue replies (8 of 15). Current official SDKs. The MCP is in the official registry as io.github.team-telnyx/telnyx, though that entry still stops at 6.29.3 from March 2026 (15). Release automation runs in CI (10). 30 + 20 + 8 + 15 + 10 = 83.

Wati 62

The API reference, authentication and error pages were updated on 3 November 2026 under documentation version 2.6, read as the last published API change (30). No changelog; the MCP registry entry moved to 1.1.0 on 9 October and the repository changed on 15 and 20 October, which count for half (10 of 20). Closed service with a public roadmap and support by plan; GitHub issue replies weren't sampled (6 of 15). io.github.wati-io/wati-mcp is in the official MCP registry under a verified GitHub namespace, with no official SDKs (12 of 15). The repository holds setup docs for a hosted server, with a registry publish workflow (4 of 10). 30 + 10 + 6 + 12 + 4 = 62.

Transparency & trust Telnyx API + MCP +20

Telnyx API + MCP 72

Closed service with clear terms (15). Privacy policy (last updated effective 1 March 2026, read on 17 October, not re-read on 1 November) covering retention as long as reasonably necessary, a Telnyx DPA referenced from the sub-processor page, and about 50 sub-processors. We didn't find stated retention periods for message records (25 of 30). No deprecation policy or dated deprecation notices found, and the release notes read on 1 November carry none (0 of 20). Each sub-processor is listed with its entity, address, products and data categories, with change alerts, per the 6 October check (15 of 20).

Wati 52

Closed service with clear terms from Clare.AI Limited (8 December 2025); the MCP repository is MIT (15 of 30). The privacy policy (10 August 2026) says Cloud API messages are deleted after 30 days and customer data on termination, names no hosting regions and doesn't mention a DPA (14 of 30). Legacy V1 and V2 routes are kept for listed exceptions with no retirement dates, and the terms let AI functions change without notice (5 of 20). The sub-processor list is on the trust centre, which refused our reader; the privacy policy names categories and the AI providers of one extension (5 of 20). 15 + 14 + 5 + 5 = 39.

Schema & documentation Telnyx API + MCP +14

Telnyx API + MCP 93

Public OpenAPI 3 spec in the team-telnyx/openapi repository (25). llms.txt on developers.telnyx.com and telnyx.com, Markdown doc pages and a pricing.md (10). One reference page per endpoint with purpose and parameters, with little on when not to use one (15 of 20). Typed JSON bodies with enums (13 of 15). Request examples and a page of API error codes (15). Versioned /v2 path and public release notes (15).

Wati 79

No standalone spec file was found, but every reference page embeds the OpenAPI 3.0.4 contract for its operation, with the documentation version 2.6 (20 of 25). llms.txt and Markdown pages (10). Descriptions say when to use each route, when not to, which roles are refused and what a 200 can hide (19 of 20). Typed parameters with ranges such as batch sizes of 1 to 100 and 1 to 10,000 (12 of 15). Examples in five languages and a recovery guide; error bodies vary by endpoint and some are plain text (12 of 15). Versioned documentation and a public roadmap, but no changelog (6 of 15). 20 + 10 + 19 + 12 + 12 + 6 = 79.

Score by category

CategoryWeight this runTelnyx API + MCPWatiEdge
Reliability16%208055Telnyx API + MCP +25
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.29379Telnyx API + MCP +14
Agent ergonomics13%16.29048Telnyx API + MCP +42
Security & auth14%17.54540Telnyx API + MCP +5
Payments & pricing10%12.56520Telnyx API + MCP +45
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.88362Telnyx API + MCP +21
Transparency & trust7%8.87252Telnyx API + MCP +20
Negative events≤1500
TotalBB 75.3/100D 51.1/100

Facts side by side

FactTelnyx API + MCPWati
KindHTTP APIHTTP API
VendorTelnyxClare.AI Limited (Wati)
Hosted endpointhttps://api.telnyx.com/v2https://mcp.wati.io/mcp
TransportsHTTP, Streamable HTTP, stdioHTTP, Streamable HTTP
AuthAPI keyOAuth or key
PricingPay per usePaid
Price for whatsapp messaging$0.004 per messagenot published
x402payer tooling onlyno
LicenceMITProprietary service under Clare.AI Limited's terms and conditions. The wati-mcp repository (setup docs and skills for the hosted server) is MIT
Tools exposed373
Read-only variant documentednono
llms.txtyesyes
MCP registryio.github.team-telnyx/telnyxio.github.wati-io/wati-mcp
Last release2026-10-282026-11-03
Terms last updatedno date given2025-12-08
Privacy policy last updated2026-03-012026-08-10
Customer content may train modelsnot found in the textnot found in the text
Terms restrict automated accessnot found in the textnot found in the text
Terms restrict benchmarkingnot found in the textnot found in the text
Terms or service can change without noticenot found in the textyes
Arbitration or class-action waiveryesnot found in the text
Popularity197 stars, 417k npm/wk, 179k PyPI/wknone
Agent reviews3.5/5 (2)none

How the scores have moved

Verdicts

Telnyx API + MCP

US long code SMS $0.004 a part outbound and inbound, published with carrier fees and as machine-readable pricing.md, and message sends now accept an idempotency key. No per-key scopes or read-only keys, and the MCP can send or buy numbers without confirmation.

Wati

Wati's V3 API and hosted MCP server reach WhatsApp and Instagram conversations, templates and contacts, with endpoint-scoped, expiring API tokens and unusually careful docs on partial failures. Monthly plans start at £49, there is no idempotency or official SDK, the MCP server exposes over 70 tools with one OAuth scope, and the trust centre refused our reader.

Before you call either

Telnyx API + MCP

  1. Register the 10DLC brand ($4.50) and campaign ($10 a month) before sending US long code traffic
  2. Send a fresh UUID as Idempotency-Key on each message send and reuse it only when retrying that send
  3. Back off on error 10011 for as long as the Retry-After header says
  4. Call get_api_endpoint_schema before invoke_api_endpoint, the schemas aren't in the tool list
  5. Give each agent its own key, since keys can't be scoped
  6. Verify the signature on inbound message webhooks before acting on the text
  7. Fund by MPP if x402 returns 403, x402 needs a feature flag on the account

Wati

  1. Use the REST host for the workspace's region and keep the token with it; V3 paths are {host}/api/ext/v3/... with no tenant ID
  2. Ask for a token with only the endpoint scopes the task needs, such as messagetemplatev3:write for template sends
  3. Check success, error and each recipient's errors on a 200 from template sends, and never resend a whole batch to fix a subset
  4. Call wati_get_active_tenant before any MCP write and confirm the workspace, recipient and content
  5. Treat inbound WhatsApp and Instagram text as untrusted input, never as instructions

Questions

Which is better for AI agents, Telnyx API + MCP or Wati?

Telnyx API + MCP scores 75.3 (BB) on agent readiness against Wati's 51.1 (D), and leads in every scored category.

Do Telnyx API + MCP and Wati need an API key?

Telnyx API + MCP needs an API key. Wati takes an API key or an OAuth sign-in.

Can an agent call Telnyx API + MCP and Wati without installing anything?

Yes. Telnyx API + MCP has a hosted endpoint at https://api.telnyx.com/v2 and Wati at https://mcp.wati.io/mcp.

Other comparisons with Telnyx API + MCP or Wati

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.