ElevenLabs Agents
Multimodal AI agents platform: build, test and deploy voice/chat agents in 70+ languages with RAG knowledge bases, external tool calls, multi-agent workflows, testing, analytics, SDKs and telephony.
Score breakdown
Task success is worth up to 80 points, word error rate and latency up to 10 each.
Testing implementation
One of the calls ElevenLabs Agents took, end to end: the simulated caller, the agent as deployed on ElevenLabs Agents, the tool server it called and the grade. Callers, tools and grading are identical on every platform; only the platform and its model stack change.
Caller
C5-H1 · Large balance, cannot pay in full
- full name
- Maria Alvarez
- zip
- 10016
- date of birth
- 1972-06-30
- mobile
- 212-555-0133
ElevenLabs Agents agent
Healthcare front desk
- STT
- Configured by the provider
- LLM
- Configured by the provider
- TTS
- Configured by the provider
- Transport
- WebSocket
- Prompt and tools
- Shared healthcare spec, 24 tools
Tool server
Bluejay healthcare systems
- Tools
- 24 HTTP tools
- State
- Seeded database per call
- Log
- Every request, keyed to the call
- Same build
- Every platform
Grade
Pass
- ✓
identify_patientfirst_name=Maria, last_name=Alvarez, dob=1972-06-30 - ✓
verify_identityfull_name=Maria Alvarez, dob=1972-06-30 - ✓
get_patient_summary - ✓
get_account_balance - ✓
explain_chargeline_item_id=li_visit - ✓
offer_financing - ✓
send_payment_linkmobile_e164=+12125550133 - ✓
find_slotslocation_ids=['loc_park_ave'] - ✓
reschedule_appointmentappointment_id=2, new_start=2026-08-24T09:00, new_end=2026-08-24T09:30
Tool-call diffempty
Database-state diffmatch
Task success by industry
Share of calls where the agent completed the caller's request. Click an industry to see what its callers ask for.
Word error rate
How well the platform heard the caller: the share of the caller's words its transcript got wrong. Lower is better. Variants replay a task with background noise or a degraded line.
Latency
How fast the agent answered: time from the caller's last word to the agent's first audio, over every turn. Full marks inside the shaded band; no points in the hatched end.
Findings
Setup review
The authors' assessment of building, deploying and running the agent on ElevenLabs Agents: interface, developer experience, customization, going live, MCP, documentation, pricing and languages. Bluejay's voice AI engineers score it during setup from documented capabilities and hands-on use. Expand a category for what we check and how each item is scored.
UIHow well someone can build, test and review an agent in the console without writing code.8 / 10
How it is scored. Six weighted line items, each graded full, half or none: time to first agent, build without code, test in the browser, call review, errors you can see, team workspace.
| What we checked | Weight | Grade | Note |
|---|---|---|---|
| Time to first agent | 2.5 | half* 1.25 | not checked |
| Build without code | 2.0 | full* 2 | Console agent builder (docs) |
| Test in the browser | 1.5 | full* 1.5 | Browser test call (docs) |
| Call review | 1.5 | full* 1.5 | Conversation history (docs) |
| Errors you can see | 1.5 | half* 0.75 | not checked |
| Team workspace | 1.0 | full* 1 | Workspaces (docs) |
Initial grades from the 28–29 Oct 2026 run and public docs; * = judged from docs, not yet verified on a call.
DevXWhether a developer can build, run and read back the whole agent from code.8.8 / 10
How it is scored. Six weighted line items: agent-as-code, call control, data out, API correctness, SDK quality, scoped keys and auth.
| What we checked | Weight | Grade | Note |
|---|---|---|---|
| Agent-as-code | 2.5 | full 2.5 | Agent updated by API, read back identical (fixes/ts_date) |
| Call control | 1.5 | full* 1.5 | Outbound/WebSocket by API (docs) |
| Data out | 2.5 | full* 2.5 | Conversation API (docs) |
| API correctness | 1.5 | half 0.75 | Dropped additionalProperties on 28 tools + ~55 keywords in healthcare (vendor_configs/REPORT.md) |
| SDK quality | 1.0 | full* 1 | Python + TS SDKs (docs) |
| Scoped keys and auth | 1.0 | half* 0.5 | not checked |
Initial grades from the 28–29 Oct 2026 run and public docs; * = judged from docs, not yet verified on a call.
Platform customizabilityHow far you can shape the agent: its models, conversation behaviour, prompt and tools.7.2 / 10
How it is scored. Seven weighted line items, including interruption handling measured on the platform's defaults during end-to-end validation.
| What we checked | Weight | Grade | Note |
|---|---|---|---|
| Model choice | 1.5 | full* 1.5 | LLM list incl. glm-52, reasoning effort settable (vendor_configs) |
| Conversation behaviour | 1.0 | full* 1 | Turn timeout/eagerness exposed (vendor_configs) |
| Interruption handling | 3.0 | half* 1.5 | Not measured |
| Prompt and context | 1.0 | full 1 | Full prompt + calendar, read back identical (ts_date) |
| Custom tools | 2.5 | half 1.25 | HTTP tools work; schema constraints dropped (vendor_configs) |
| Speech controls | 0.5 | full* 0.5 | Voice settings exposed |
| Agent handoff | 0.5 | full* 0.5 | Agent transfer (docs) |
Initial grades from the 28–29 Oct 2026 run and public docs; * = judged from docs, not yet verified on a call.
Go liveEverything needed to take the agent to production and keep it running.4.2 / 10
How it is scored. Seven weighted line items, including runtime under load: the break rate at 20 concurrent calls against the 2-call baseline.
| What we checked | Weight | Grade | Note |
|---|---|---|---|
| Telephony and numbers | 2.0 | half* 1 | Twilio/SIP integration (docs) |
| Release management | 1.5 | half* 0.75 | not checked |
| Runtime under load | 2.5 | none 0 | NO_ANSWER burst at 14 (cut to 7); quota cap cut calls mid-call (concurrency_change.jsonl) |
| Integrations at the edge | 1.0 | half* 0.5 | not checked |
| Ops and incident response | 1.5 | half 0.75 | Quota hit surfaced only as mid-call cutoffs / NO_ANSWER |
| Compliance and trust | 1.0 | full* 1 | SOC 2 / HIPAA claimed (docs) |
| Support and escalation | 0.5 | half* 0.25 | not checked |
Initial grades from the 28–29 Oct 2026 run and public docs; * = judged from docs, not yet verified on a call.
MCPWhether the agent can use tools over MCP, and whether coding agents can build and run the platform through MCP.8.2 / 10
How it is scored. Four weighted line items: MCP tools on a live call, connection controls, a platform MCP server, and how much of the API that server covers.
| What we checked | Weight | Grade | Note |
|---|---|---|---|
| MCP tools on a live call | 3.5 | full* 3.5 | MCP servers on agents (docs) |
| MCP connection controls | 1.5 | half* 0.75 | Approval modes (docs, unverified) |
| Platform MCP server | 3.0 | full* 3 | Hosted MCP server manages agents (elevenlabs.io/docs/eleven-agents/operate/hosted-mcp) |
| What that server can do | 2.0 | half* 1 | Mostly TTS/voice; agent coverage not measured |
Initial grades from the 28–29 Oct 2026 run and public docs; * = judged from docs, not yet verified on a call.
DocsWhether the docs are accurate, complete and usable by people and by coding agents.5.5 / 10
How it is scored. Eight weighted line items, from reference accuracy and payload schemas to change management and machine-readability.
| What we checked | Weight | Grade | Note |
|---|---|---|---|
| Reference accuracy | 2.0 | half 1 | No documented default LLM; similarity_boost/speculative_turn differ from docs (vendor_configs) |
| Payload and event schemas | 1.5 | half* 0.75 | not checked |
| Concepts and lifecycle | 1.0 | half* 0.5 | not checked |
| Telephony and go-live guides | 1.5 | half* 0.75 | not checked |
| Errors, limits and gotchas | 1.0 | half* 0.5 | not checked |
| Examples and quickstarts | 1.0 | half* 0.5 | not checked |
| Change management | 1.0 | half* 0.5 | not checked |
| Machine-readability | 1.0 | full* 1 | llms.txt; OpenAPI not verified |
Initial grades from the 28–29 Oct 2026 run and public docs; * = judged from docs, not yet verified on a call.
Pricing clarityWhether a buyer can predict what the platform will cost and see what it is costing.5.5 / 10
How it is scored. Six weighted line items, including cost per hour: our estimate from public pricing against the real bill for the scored calls.
| What we checked | Weight | Grade | Note |
|---|---|---|---|
| Public price | 2.0 | full* 2 | Public plans |
| Every line item named | 1.5 | half* 0.75 | $0.08/min; LLM separate, STT/TTS bundled (elevenlabs.io/pricing/agents) |
| Cost per hour matches the estimate | 3.0 | half* 1.5 | Bill not reconciled |
| Usage and spend in the product | 1.5 | half* 0.75 | not checked |
| Spend controls | 1.5 | none 0 | Usage cap cut calls mid-call ("This request exceeds your quota limit.") and NO_ANSWER'd new calls; no graceful stop |
| Free start | 0.5 | full* 0.5 | Free tier |
Initial grades from the 28–29 Oct 2026 run and public docs; * = judged from docs, not yet verified on a call.
Language supportWhether the agent can take a call in Spanish, French, Mandarin Chinese, Hindi or Arabic, understand it and reply in that language.10 / 10
How it is scored. One translated probe task per language, two runs each. Full when the agent understood and replied in the language, half when it understood but replied in English.
| What we checked | Weight | Grade | Note |
|---|---|---|---|
| Spanish | 2.0 | full* 2 | Graded by configuration; live probes not run |
| French | 2.0 | full* 2 | Graded by configuration; live probes not run |
| Mandarin Chinese | 2.0 | full* 2 | Graded by configuration; live probes not run |
| Hindi | 2.0 | full* 2 | Graded by configuration; live probes not run |
| Arabic | 2.0 | full* 2 | Graded by configuration; live probes not run |
Initial grades from the 28–29 Oct 2026 run and public docs; * = judged from docs, not yet verified on a call.
Run details
Model configuration
- STT
- Configured by the provider
- LLM
- Configured by the provider
- TTS
- Configured by the provider
- Transport
- WebSocket
Run
- Scored calls
- 630
- Runs per task
- 5
- Run dates
- 21–22 Oct 2026
- Concurrency held
- up to 14 calls
- Turns measured
- 2,952 turns