Rooms Participants And Tracks
LiveKit (Cloud + Agents) · LiveKit
Real-time Voice & Video Infra — LiveKit
Evaluates LiveKit's Rooms, Participants & Tracks across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Real-time Voice & Video Infra eval coverage.
About LiveKit
LiveKit is open-source real-time voice/video infrastructure used to build voice agents and live experiences — a WebRTC SFU, telephony (SIP), recording/egress, and the LiveKit Agents framework for STT→LLM→TTS pipelines, available as LiveKit Cloud and self-hosted.
Sample tests· showing 3 of 9
| # | Input | Expected behavior | Check |
|---|---|---|---|
| 01 | Two browser tabs of the same user join the same LiveKit room with identity='user-42'. The second join arrives while the first is still publishing audio. | Per LiveKit docs, participant identity is unique per room — the new join displaces the prior session. Integrator must detect the kicked-out event on the first client (disconnect with reason DUPLICATE_IDENTITY) and decide between (a) surface 'opened in another tab' UI, or (b) namespace identity per … | Pass / FailAi Platformhigh |
| 02 | Operator stores JSON in Room.metadata: '{"tenant":"acme","plan":"pro"}'. A client SDK reads room.metadata and expects an object. | Per docs, Room.metadata is an opaque string up to the published size cap. Clients receive a string and must JSON.parse client-side. Integrator should version the schema (e.g., {"v":1,...}) and tolerate unknown fields. Do not rely on the SDK to deserialize for you. | Pass / FailAi Platformmedium |
| 03 | CreateRoom sets empty_timeout=300 (5 min) and departure_timeout=20. After the last participant leaves, agent expects the room to live another 5 minutes for graceful reconnect. | empty_timeout governs lifetime when the room never had participants; departure_timeout governs lifetime after the last participant leaves. Choose both deliberately — short departure_timeout drops rooms during transient client reconnects (mobile network handover); long departure_timeout wastes resou… | Pass / FailAi Platformmedium |
How this eval is graded
Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
Rubric criteria
- Livekit
- Ai Platform
- Rooms Participants And Tracks
Recommended for
Works with
Related evals
Claude API
Evaluates Anthropic's Batch API across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Extended Thinking across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Files API & Citations across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
ViewFrequently asked questions
What does the Rooms Participants And Tracks eval for LiveKit LiveKit (Cloud + Agents) test?+
Evaluates LiveKit's Rooms, Participants & Tracks across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Real-time Voice & Video Infra eval coverage.
How is the Rooms Participants And Tracks eval scored?+
The judge rubric: Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
How many test cases does this eval pack include?+
The Rooms Participants And Tracks pack for LiveKit LiveKit (Cloud + Agents) contains 9 test cases. 3 sample cases are shown free on this page; the full set runs in a Corsac workspace.
How do I run this eval?+
Sign up for Corsac, connect your model or agent endpoint, and run the Rooms Participants And Tracks pack as-is or after customizing thresholds. Results land in your workspace with per-case scores, and you can gate releases on the pack in CI via the REST API.
Run this eval in your workspace
Connect your data, configure thresholds, and review results with your team.