Refines default behavior for preemptive generation to better handle long or intermittent user speech, reducing unnecessary downstream inference and associated cost increases.

Also introduces PreemptiveGenerationOptions for developers who need fine-grained control over this behavior.

https://github.com/livekit/agents/blob/78a66bcf79c5cea82989401c408f1dff4b961a5b/livekit-agents/livekit/agents/voice/turn.py#L115

class PreemptiveGenerationOptions(TypedDict, total=False):
    """Configuration for preemptive generation."""

    enabled: bool
    """Whether preemptive generation is enabled. Defaults to ``True``."""

    preemptive_tts: bool
    """Whether to also run TTS preemptively before the turn is confirmed.
    When ``False`` (default), only LLM runs preemptively; TTS starts once the
    turn is confirmed and the speech is scheduled."""

    max_speech_duration: float
    """Maximum user speech duration (s) for which preemptive generation
    is attempted. Beyond this threshold, preemptive generation is skipped
    since long utterances are more likely to change and users may expect
    slower responses. Defaults to ``10.0``."""

    max_retries: int
    """Maximum number of preemptive generation attempts per user turn.
    The counter resets when the turn completes. Defaults to ``3``."""

What's Changed

fix(voice): add PreemptiveGenerationOptions for fine-grained control by @longcw in https://github.com/livekit/agents/pull/5428

Full Changelog: https://github.com/livekit/agents/compare/livekit-agents@1.5.3...livekit-agents@1.5.4

livekit-agents@1.5.3

↗

What's Changed

feat(amd): add OTEL span and tag for AMD by @chenghao-mou in https://github.com/livekit/agents/pull/5376
fix(openai): prepend session instructions in realtime generate_reply by @longcw in https://github.com/livekit/agents/pull/5394
fix: AgentTask deadlock when on_enter awaits generate_reply that triggers another AgentTask by @longcw in https://github.com/livekit/agents/pull/5377
telemetry: emit OTel span events for developer-role messages by @joaquinhuigomez in https://github.com/livekit/agents/pull/5403
feat(realtime): reuse realtime session across agent handoffs if supported by @longcw in https://github.com/livekit/agents/pull/5229
fix(llm): handle double-encoded JSON tool arguments from providers by @prettyprettyprettygood in https://github.com/livekit/agents/pull/5409
chore: exposed the max session duration in with_azure() function by @k-1208 in https://github.com/livekit/agents/pull/5383
(gemini live 3.1): fix tool responses by @tinalenguyen in https://github.com/livekit/agents/pull/5413
fix(google): capture usage_metadata before early continues in streaming by @Panmax in https://github.com/livekit/agents/pull/5404
fix(voice): block new user turns immediately on update_agent() to prevent transition delay by @Panmax in https://github.com/livekit/agents/pull/5396
fix(tests): update drive thru instructions by @chenghao-mou in https://github.com/livekit/agents/pull/5405
feat(inference): handle preflight_transcript in inference STT plugin by @adrian-cowham in https://github.com/livekit/agents/pull/5412
fix(aws): unwrap doubly-encoded JSON tool arguments from Nova Sonic by @rililinx in https://github.com/livekit/agents/pull/5411
chore: pin GHA by commit by @davidzhao in https://github.com/livekit/agents/pull/5415
chore(deps): update dependency langchain-core to v1.2.28 [security] by @renovate[bot] in https://github.com/livekit/agents/pull/5417
chore(deps): update dependency aiohttp to v3.13.4 [security] by @renovate[bot] in https://github.com/livekit/agents/pull/5416
chore(deps): update dependency nltk to v3.9.4 [security] by @renovate[bot] in https://github.com/livekit/agents/pull/5418
(azure openai): ensure gpt-realtime-1.5 compatibility by @tinalenguyen in https://github.com/livekit/agents/pull/5407
chore(deps): update github workflows (major) by @renovate[bot] in https://github.com/livekit/agents/pull/5424
update: Sarvam STT - add verbose error loggin and remove retry connection by @dhruvladia-sarvam in https://github.com/livekit/agents/pull/5373
fix(inworld): do not leak connections when when cancelled by @davidzhao in https://github.com/livekit/agents/pull/5427
feat: add service_tier parameter to Responses API LLM by @piyush-gambhir in https://github.com/livekit/agents/pull/5346
Feature/krisp viva sdk support by @realgarik in https://github.com/livekit/agents/pull/4370
fix: empty transcript blocks commit_user_turn until timeout by @longcw in https://github.com/livekit/agents/pull/5429
fix: allow multiple AsyncToolsets by deduplicating management tools by @longcw in https://github.com/livekit/agents/pull/5369
feat(beta/workflows): add InstructionParts for modular instruction customization by @longcw in https://github.com/livekit/agents/pull/5077
add ToolSearchToolset and ToolProxyToolset for dynamic tool discovery by @longcw in https://github.com/livekit/agents/pull/5140
Feature - Configurable session close transcript timeout by @bml1g12 in https://github.com/livekit/agents/pull/5328
Fix FrameProcessor lifecycle for selector based noise cancellation by @Topherhindman in https://github.com/livekit/agents/pull/5433
feat: add Runway Characters avatar plugin by @robinandeer in https://github.com/livekit/agents/pull/5355
Rename e2ee to encryption in JobContext.connect by @longcw in https://github.com/livekit/agents/pull/5454
chore: reduce renovate noise by @davidzhao in https://github.com/livekit/agents/pull/5421
fix(liveavatar): wait for connected state and chunk audio before sending by @dyi1 in https://github.com/livekit/agents/pull/5453
(phonic): support realtimemodel say() by @tinalenguyen in https://github.com/livekit/agents/pull/5293
feat: add Cerebras LLM plugin by @u9g in https://github.com/livekit/agents/pull/5456
(google tts): add "gemini-3.1-flash-tts-preview" model by @tinalenguyen in https://github.com/livekit/agents/pull/5459
(hedra) remove from examples and raise exception by @tinalenguyen in https://github.com/livekit/agents/pull/5460
xai stt by @tinalenguyen in https://github.com/livekit/agents/pull/5458
feat(openai): expose max_output_tokens on Responses API LLM by @piyush-gambhir in https://github.com/livekit/agents/pull/5449
(xai stt): pass diarization capability + minor fix by @tinalenguyen in https://github.com/livekit/agents/pull/5461
Adding xAI Grok llm support for inference by @russellmartin-livekit in https://github.com/livekit/agents/pull/5201
(release workflow): allow spacing in dependency by @tinalenguyen in https://github.com/livekit/agents/pull/5463
livekit-agents@1.5.3 by @github-actions[bot] in https://github.com/livekit/agents/pull/5464

New Contributors

@joaquinhuigomez made their first contribution in https://github.com/livekit/agents/pull/5403
@k-1208 made their first contribution in https://github.com/livekit/agents/pull/5383
@rililinx made their first contribution in https://github.com/livekit/agents/pull/5411
@renovate[bot] made their first contribution in https://github.com/livekit/agents/pull/5417
@realgarik made their first contribution in https://github.com/livekit/agents/pull/4370
@robinandeer made their first contribution in https://github.com/livekit/agents/pull/5355
@dyi1 made their first contribution in https://github.com/livekit/agents/pull/5453
@u9g made their first contribution in https://github.com/livekit/agents/pull/5456

Full Changelog: https://github.com/livekit/agents/compare/livekit-agents@1.5.2...livekit-agents@1.5.3

Apr 8, 2026

livekit-agents@1.5.2

↗

Note

livekit-agents 1.5 introduced many new features. You can check out the changelog here.

What's Changed

Update Phonic generate_reply timeout to 10 seconds by @qionghuang6 in https://github.com/livekit/agents/pull/5205
fix: pass prometheus_multiproc_dir in from_server_options initialization by @ivanbalingit in https://github.com/livekit/agents/pull/5195
feat(mistralai): upgrade to SDK v2 by @Pauldevillers in https://github.com/livekit/agents/pull/5163
(deepgram sttv2): validate eager_eot_threshold value by @tinalenguyen in https://github.com/livekit/agents/pull/5216
Add WebSocket streaming support to Baseten TTS plugin by @iancarrasco-b10 in https://github.com/livekit/agents/pull/4741
fix: allow codec format specification via the user for Sarvam TTS by @pUrGe12 in https://github.com/livekit/agents/pull/5209
emit agent handoff from conversation_item_added by @tinalenguyen in https://github.com/livekit/agents/pull/5218
fix(llm): surface validation error details to LLM on function call argument failures by @Lyt060814 in https://github.com/livekit/agents/pull/5193
fix(cli): update log level width in console mode by @chenghao-mou in https://github.com/livekit/agents/pull/5224
fix(utils): preserve type annotations in deprecate_params by @longcw in https://github.com/livekit/agents/pull/5200
fix(test): replace oai with deepgram and fix broken tests by @chenghao-mou in https://github.com/livekit/agents/pull/5225
feat(voice): reuse STT connection across agent handoffs by @longcw in https://github.com/livekit/agents/pull/5093
feat(google): add VertexRAGRetrieval provider tool by @youpesh in https://github.com/livekit/agents/pull/5222
fix: ensure MCP client enter/exit run in the same task by @longcw in https://github.com/livekit/agents/pull/5223
feat(assemblyai): add domain parameter for Medical Mode by @m-ods in https://github.com/livekit/agents/pull/5208
fix: Nova Sonic interactive context bugs and dynamic tool support by @prettyprettyprettygood in https://github.com/livekit/agents/pull/5220
(google realtime): add gemini-3.1-flash-live-preview model by @tinalenguyen in https://github.com/livekit/agents/pull/5233
fix(utils): improve type annotation for deprecate_params decorator by @longcw in https://github.com/livekit/agents/pull/5244
fix: expose endpointing_opts in AgentSession.update_options() by @longcw in https://github.com/livekit/agents/pull/5243
Fix/stt fallback adapter propagate aligned transcript by @miladmnasr in https://github.com/livekit/agents/pull/5237
feat(mistral): add voxtral TTS support by @jeanprbt in https://github.com/livekit/agents/pull/5245
feat(anthropic): support strict tool use schema by @roshan-shaik-ml in https://github.com/livekit/agents/pull/5259
Baseten Plugin Update: fix metadata schema, add chain_id support, and improve response parsing by @jiegong-fde in https://github.com/livekit/agents/pull/4889
feat(upliftai): add support for phrase replacement config id by @zaidqureshi2 in https://github.com/livekit/agents/pull/5261
feat(soniox): expose max_endpoint_delay_ms option by @pstrav in https://github.com/livekit/agents/pull/5214
fix: prevent TTS retry after partial audio and replay input on retry by @longcw in https://github.com/livekit/agents/pull/5242
fix: only start session host when it's primary session by @longcw in https://github.com/livekit/agents/pull/5241
fix: prevent CancelledError from propagating to unrelated Tee peers by @longcw in https://github.com/livekit/agents/pull/5273
fix: prevent AttributeError in ThreadJobExecutor.logging_extra() by @longcw in https://github.com/livekit/agents/pull/5277
fix(openai): close current generation channels on realtime reconnect by @longcw in https://github.com/livekit/agents/pull/5276
fix(recorder): guard against empty agent speech frames by @chenghao-mou in https://github.com/livekit/agents/pull/5279
fix(stt): reset VAD when STT sends EOT by @chenghao-mou in https://github.com/livekit/agents/pull/5095
feat(anam): add avatarModel config support by @sr-anam in https://github.com/livekit/agents/pull/5272
fix: catch TimeoutError from drain() so aclose() always runs by @seglo in https://github.com/livekit/agents/pull/5282
(gemini-3.1-flash-live-preview): add warning for generate_reply by @tinalenguyen in https://github.com/livekit/agents/pull/5286
feat(mistralai): add ref_audio support to Voxtral TTS for zero-shot voice cloning by @EtienneLescot in https://github.com/livekit/agents/pull/5278
fix(core): reset user state to listening when audio is disabled by @chenghao-mou in https://github.com/livekit/agents/pull/5198
append generate_reply instructions as system msg and convert it to user msg if unsupported by @longcw in https://github.com/livekit/agents/pull/5287
add AsyncToolset by @longcw in https://github.com/livekit/agents/pull/5127
fix(core): fix BackgroundAudioPlayer.play() hanging indefinitely by @theomonnom in https://github.com/livekit/agents/pull/5299
fix(cli): prevent api_key/api_secret from leaking in tracebacks by @theomonnom in https://github.com/livekit/agents/pull/5300
(phonic) Update languages fields by @qionghuang6 in https://github.com/livekit/agents/pull/5285
fix(core): reduce TTS output buffering latency by @theomonnom in https://github.com/livekit/agents/pull/5292
add session_end_timeout and gracefully cancel entrypoint on shutdown by @theomonnom in https://github.com/livekit/agents/pull/4580
feat: OTEL metrics for latencies, usage, and connection timing by @theomonnom in https://github.com/livekit/agents/pull/4891
evals: custom judges, tag metadata, and OTEL improvements by @theomonnom in https://github.com/livekit/agents/pull/5306
fix is_context_type for generic RunContext types by @theomonnom in https://github.com/livekit/agents/pull/5307
add 7-day uv cooldown by @chenghao-mou in https://github.com/livekit/agents/pull/5290
fix(openai realtime): support per-response tool_choice in realtime sessions by @longcw in https://github.com/livekit/agents/pull/5211
use delta aggregation temporality for otel metrics by @paulwe in https://github.com/livekit/agents/pull/5314
(phonic) Add min_words_to_interrupt to Phonic plugin options by @qionghuang6 in https://github.com/livekit/agents/pull/5304
add tag field to evaluation OTEL log records by @theomonnom in https://github.com/livekit/agents/pull/5315
docs: add example agent replies to AsyncToolset by @longcw in https://github.com/livekit/agents/pull/5313
fix(cartesia): handle flush_done message in TTS _recv_task by @Panmax in https://github.com/livekit/agents/pull/5321
fix(voice): make function call history preservation configurable in AgentTask by @GopalGB in https://github.com/livekit/agents/pull/5288
fix: convert oneOf to anyOf in strict schema for discriminated unions by @longcw in https://github.com/livekit/agents/pull/5324
(gemini realtime): add warnings in update_chat_ctx and update_instructions by @tinalenguyen in https://github.com/livekit/agents/pull/5332
fix: wait_for_participant waits until participant is fully active by @davidzhao in https://github.com/livekit/agents/pull/5271
feat: answering machine detection by @chenghao-mou in https://github.com/livekit/agents/pull/4906
feat: expose service_tier in CompletionUsage from OpenAI Responses API by @piyush-gambhir in https://github.com/livekit/agents/pull/5341
fix: add PARTICIPANT_KIND_CONNECTOR to default participant kinds by @anunaym14 in https://github.com/livekit/agents/pull/5339
feat/sarvam-llm-openai-compatible-integration by @dhruvladia-sarvam in https://github.com/livekit/agents/pull/5069
feat(azure-stt): Possibility to change segmentation options during a call by @rafallezanko in https://github.com/livekit/agents/pull/5323
fix(sarvam): sync missing API params, fix value ranges, and update models by @Namit1867 in https://github.com/livekit/agents/pull/5347
(xai tts): update fields and ws setup by @tinalenguyen in https://github.com/livekit/agents/pull/5350
fix(smallestai): add lightning-v3.1 endpoint routing by @sg-siddhant in https://github.com/livekit/agents/pull/5330
feat(inference): add debug/identification headers to inference requests by @adrian-cowham in https://github.com/livekit/agents/pull/5337
Move community plugins to livekit-plugins/community/ by @theomonnom in https://github.com/livekit/agents/pull/5250
feat: support per-response tools in generate_reply by @longcw in https://github.com/livekit/agents/pull/5310
fix xAI realtime update chat ctx by @longcw in https://github.com/livekit/agents/pull/5320
Fix RoomIO teardown listener cleanup by @sindarknave in https://github.com/livekit/agents/pull/5357
feat(mistral): support voxtral realtime streaming stt & modernize mistral plugin by @jeanprbt in https://github.com/livekit/agents/pull/5289
fix: say() with missing audio file hangs forever and blocks speech queue by @theomonnom in https://github.com/livekit/agents/pull/5358
add prompt_cache_retention chat completion option to inference by @s-hamdananwar in https://github.com/livekit/agents/pull/5370
Add Murf as optional dep by @royalfig in https://github.com/livekit/agents/pull/5334
feat(core): Support multiple provider keys in extra_content serialization by @adrian-cowham in https://github.com/livekit/agents/pull/5374
ci: add PyPI publish workflow with trusted publishing by @theomonnom in https://github.com/livekit/agents/pull/5379
feat: Add D-ID avatar plugin by @osimhi213 in https://github.com/livekit/agents/pull/5232
ci: fix tag checkout and discover glob by @theomonnom in https://github.com/livekit/agents/pull/5381
feat(rime): add mistv3 model support by @mcullan in https://github.com/livekit/agents/pull/5298
ci: fix update_versions.py invocation by @theomonnom in https://github.com/livekit/agents/pull/5382
ci: remove release label from publish workflow by @theomonnom in https://github.com/livekit/agents/pull/5384
require livekit-protocol>=1.1.5, implement get_framework_info by @theomonnom in https://github.com/livekit/agents/pull/5385
ci: fix build permissions and tag format by @theomonnom in https://github.com/livekit/agents/pull/5386
ci: fix version read in publish workflow by @theomonnom in https://github.com/livekit/agents/pull/5388
ci: use livekit-agents@version for release PR title by @theomonnom in https://github.com/livekit/agents/pull/5390
fix: minimax optional dep not bumped by update_versions.py by @theomonnom in https://github.com/livekit/agents/pull/5392
livekit-agents@1.5.2 by @github-actions[bot] in https://github.com/livekit/agents/pull/5391

New Contributors

@ivanbalingit made their first contribution in https://github.com/livekit/agents/pull/5195
@Pauldevillers made their first contribution in https://github.com/livekit/agents/pull/5163
@iancarrasco-b10 made their first contribution in https://github.com/livekit/agents/pull/4741
@pUrGe12 made their first contribution in https://github.com/livekit/agents/pull/5209
@Lyt060814 made their first contribution in https://github.com/livekit/agents/pull/5193
@youpesh made their first contribution in https://github.com/livekit/agents/pull/5222
@m-ods made their first contribution in https://github.com/livekit/agents/pull/5208
@prettyprettyprettygood made their first contribution in https://github.com/livekit/agents/pull/5220
@miladmnasr made their first contribution in https://github.com/livekit/agents/pull/5237
@jeanprbt made their first contribution in https://github.com/livekit/agents/pull/5245
@roshan-shaik-ml made their first contribution in https://github.com/livekit/agents/pull/5259
@jiegong-fde made their first contribution in https://github.com/livekit/agents/pull/4889
@pstrav made their first contribution in https://github.com/livekit/agents/pull/5214
@sr-anam made their first contribution in https://github.com/livekit/agents/pull/5272
@seglo made their first contribution in https://github.com/livekit/agents/pull/5282
@EtienneLescot made their first contribution in https://github.com/livekit/agents/pull/5278
@GopalGB made their first contribution in https://github.com/livekit/agents/pull/5288
@piyush-gambhir made their first contribution in https://github.com/livekit/agents/pull/5341
@anunaym14 made their first contribution in https://github.com/livekit/agents/pull/5339
@Namit1867 made their first contribution in https://github.com/livekit/agents/pull/5347
@sg-siddhant made their first contribution in https://github.com/livekit/agents/pull/5330
@sindarknave made their first contribution in https://github.com/livekit/agents/pull/5357
@royalfig made their first contribution in https://github.com/livekit/agents/pull/5334
@osimhi213 made their first contribution in https://github.com/livekit/agents/pull/5232
@mcullan made their first contribution in https://github.com/livekit/agents/pull/5298

Full Changelog: https://github.com/livekit/agents/compare/livekit-agents@1.5.1...livekit-agents@1.5.2

Mar 23, 2026

livekit-agents@1.5.1

↗

Note

livekit-agents 1.5 introduced many new features. You can check out the changelog here.

What's Changed

fix azure openai realtime support & add realtime models tests by @theomonnom in https://github.com/livekit/agents/pull/5168
fix(core): version mismatch due to bad merge by @chenghao-mou in https://github.com/livekit/agents/pull/5176
fix(turn-detector): relax transformers upper bound to allow 5.x by @gdoermann in https://github.com/livekit/agents/pull/5174
(gladia & soniox): add translation support by @tinalenguyen in https://github.com/livekit/agents/pull/5148
feat(agents): support LIVEKIT_OBSERVABILITY_URL for custom observability endpoints by @theomonnom in https://github.com/livekit/agents/pull/5179
(xai tts): update websocket endpoint by @tinalenguyen in https://github.com/livekit/agents/pull/5180
fix(core): restore chat topic support in room IO by @chenghao-mou in https://github.com/livekit/agents/pull/5181
Unskip Tool Call Items before Summarization in Task Group by @toubatbrian in https://github.com/livekit/agents/pull/5169
add sdk_version to SessionReport for observability by @theomonnom in https://github.com/livekit/agents/pull/5182
feat(hamming): add hamming monitoring plugin package by @duchammingai in https://github.com/livekit/agents/pull/5135
chore(mypy): enable mypy cache in type checking by @chenghao-mou in https://github.com/livekit/agents/pull/5192
fix: expose Chirp 3 google STT endpoint sensitivity by @karlsonlee-livekit in https://github.com/livekit/agents/pull/5196
add MCPToolset by @longcw in https://github.com/livekit/agents/pull/5138
Feat/personaplex plugin by @milanperovic in https://github.com/livekit/agents/pull/4660
fix: skip redundant realtime events in OpenAI plugin by @theomonnom in https://github.com/livekit/agents/pull/5204
feat: enable AGC by default on RoomInput audio by @theomonnom in https://github.com/livekit/agents/pull/5185
bump minimum livekit sdk version to 1.1.3 by @theomonnom in https://github.com/livekit/agents/pull/5206
livekit-agents 1.5.1 by @theomonnom in https://github.com/livekit/agents/pull/5207

New Contributors

@duchammingai made their first contribution in https://github.com/livekit/agents/pull/5135
@karlsonlee-livekit made their first contribution in https://github.com/livekit/agents/pull/5196
@milanperovic made their first contribution in https://github.com/livekit/agents/pull/4660

Full Changelog: https://github.com/livekit/agents/compare/livekit-agents@1.5.0...livekit-agents@1.5.1

Mar 19, 2026

The headline feature of v1.5.0: an audio-based ML model that distinguishes genuine user interruptions from incidental sounds like backchannels ("mm-hmm"), coughs, sighs, or background noise. Enabled by default — no configuration needed.

Key stats:

86% precision and 100% recall at 500ms overlapping speech
Rejects 51% of traditional VAD false positives
Detects true interruptions 64% faster than VAD alone
Inference completes in 30ms or less

When a false interruption is detected, the agent automatically resumes playback from where it left off — no re-generation needed.

To opt out and use VAD-only interruption:

session = AgentSession(
    ...
    turn_handling=TurnHandlingOptions(
        interruption={
            "mode": "vad",
        },
    ),
)

Blog post: https://livekit.com/blog/adaptive-interruption-handling

Dynamic Endpointing

Endpointing delays now adapt to each conversation's natural rhythm. Instead of a fixed silence threshold, the agent uses an exponential moving average of pause durations to dynamically adjust when it considers the user's turn complete.

session = AgentSession(
    ...
    turn_handling=TurnHandlingOptions(
        endpointing={
            "mode": "dynamic",
            "min_delay": 0.3,
            "max_delay": 3.0,
        },
    ),
)

New `TurnHandlingOptions` API

Endpointing and interruption settings are now consolidated into a single TurnHandlingOptions dict passed to AgentSession. Old keyword arguments (min_endpointing_delay, allow_interruptions, etc.) still work but are deprecated and will emit warnings.

session = AgentSession(
    turn_handling={
        "turn_detection": "vad",
        "endpointing": {"min_delay": 0.5, "max_delay": 3.0},
        "interruption": {"enabled": True, "mode": "adaptive"},
    },
)

Session Usage Tracking

New SessionUsageUpdatedEvent provides structured, per-model usage data — token counts, character counts, and audio durations — broken down by provider and model:

@session.on("session_usage_updated")
def on_usage(ev: SessionUsageUpdatedEvent):
    for usage in ev.usage.model_usage:
        print(f"{usage.provider}/{usage.model}: {usage}")

Usage types: LLMModelUsage, TTSModelUsage, STTModelUsage, InterruptionModelUsage.

You can also access aggregated usage at any time via the session.usage property:

usage = session.usage
for model_usage in usage.model_usage:
    print(model_usage)

Usage data is also included in SessionReport (via model_usage), so it's available in post-session telemetry and reporting out of the box.

Per-Turn Latency on `ChatMessage.metrics`

Each ChatMessage now carries a metrics field (MetricsReport) with per-turn latency data:

transcription_delay — time to obtain transcript after end of speech
end_of_turn_delay — time between end of speech and turn decision
on_user_turn_completed_delay — time in the developer callback

Action-Aware Chat Context Summarization

Context summarization now includes function calls and their outputs when building summaries, preserving tool-use context across the conversation window.

Configurable Log Level

Set the agent log level via LIVEKIT_LOG_LEVEL environment variable or through ServerOptions, without touching your code.

Deprecations

Deprecated	Replacement	Notes
`metrics_collected` event	`session_usage_updated` event + `ChatMessage.metrics`	Usage/cost data moves to `session_usage_updated`; per-turn latency moves to `ChatMessage.metrics`. Old listeners still work with a deprecation warning.
`UsageCollector`	`ModelUsageCollector`	New collector supports per-model/provider breakdown
`UsageSummary`	`LLMModelUsage`, `TTSModelUsage`, `STTModelUsage`	Typed per-service usage classes
`RealtimeModelBeta`	`RealtimeModel`	Beta API removed
`AgentFalseInterruptionEvent.message` / `.extra_instructions`	Automatic resume via adaptive interruption	Accessing these fields logs a deprecation warning
`AgentSession` kwargs: `min_endpointing_delay`, `max_endpointing_delay`, `allow_interruptions`, `discard_audio_if_uninterruptible`, `min_interruption_duration`, `min_interruption_words`, `turn_detection`, `false_interruption_timeout`, `resume_false_interruption`	`turn_handling=TurnHandlingOptions(...)`	Old kwargs still work but emit deprecation warnings. Will be removed in v2.0.
`Agent` / `AgentTask` kwargs: `turn_detection`, `min_endpointing_delay`, `max_endpointing_delay`, `allow_interruptions`	`turn_handling=TurnHandlingOptions(...)`	Same migration path as `AgentSession`. Will be removed in future versions.