> ## Documentation Index > Fetch the complete documentation index at: https://docs.liveavatar.com/llms.txt > Use this file to discover all available pages before exploring further. # Create FULL Session Token > Create session token for an API-key-authenticated user. Minting a session token is only doable via an API key (``APIUserDep``) — a session JWT (``SessionUser``) cannot mint a token for another participant. HeyGen gateway identities mint through ``/heygen_session_token``. This route only authenticates native API keys. > [!TIP] > > Bringing your own agent? [Create a LITE session token](https://docs.liveavatar.com/api-reference/sessions/create-session-token.md) instead. ## OpenAPI ````yaml /openapi-full.json post /v1/sessions/token openapi: 3.1.0 info: title: Live avatar API version: 0.1.0 servers: - url: https://api.liveavatar.com security: [] paths: /v1/sessions/token: post: tags: - Sessions summary: Create Session Token description: >- Create session token for an API-key-authenticated user. Minting a session token is only doable via an API key (``APIUserDep``) — a session JWT (``SessionUser``) cannot mint a token for another participant. HeyGen gateway identities mint through ``/heygen_session_token``. This route only authenticates native API keys. operationId: create_session_token_v1_sessions_token_post requestBody: content: application/json: schema: $ref: '#/components/schemas/FullSDKSessionTokenConfigDataSchema' required: true responses: '200': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/Response_SDKSessionTokenSchema_' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' security: - APIKeyHeader: [] components: schemas: FullSDKSessionTokenConfigDataSchema: properties: mode: type: string const: FULL title: Mode default: FULL avatar_id: type: string format: uuid title: Avatar Id is_sandbox: type: boolean title: Is Sandbox default: false video_settings: anyOf: - $ref: '#/components/schemas/VideoSettings' - type: 'null' description: Video encoding settings max_session_duration: title: Max Session Duration description: >- Maximum session duration in seconds. Must be <= the configured limit for your subscription tier. type: integer avatar_persona: anyOf: - $ref: '#/components/schemas/AvatarPersonaSchema' - type: 'null' description: >- Inline voice / context / voice_settings. Mutually exclusive with voice_agent. deprecated: true voice_agent: anyOf: - $ref: '#/components/schemas/VoiceAgentRef' - type: 'null' description: >- Reference a stored voice_agent by id; the server resolves its agent_type and builds the session accordingly (which per-session overrides apply depends on the type — see VoiceAgentRef). An elevenlabs_agent produces a LITE session even on this FULL request. Mutually exclusive with avatar_persona. interactivity_type: $ref: '#/components/schemas/SessionInteractivityTypeEnum' default: CONVERSATIONAL llm_configuration_id: title: Llm Configuration Id description: Optional custom LLM configuration ID type: string format: uuid dynamic_variables: title: Dynamic Variables description: >- Values for ${var} placeholders in the context's opening_text and prompt. Required keys are derived from the context; extra keys are ignored. At most 50 entries; keys <= 64 chars; values <= 1000 chars. additionalProperties: type: string type: object memory: anyOf: - $ref: '#/components/schemas/MemoryRef' - type: 'null' description: >- Attach session memory. Provide `prev_session_id` to seed/reuse from a prior session, `session_memory_id` to attach an existing memory, or both (must resolve to the same memory_id). type: object required: - avatar_id title: Full Mode Response_SDKSessionTokenSchema_: properties: code: type: integer title: Code default: 100 data: anyOf: - $ref: '#/components/schemas/SDKSessionTokenSchema' - type: 'null' message: title: Message type: string type: object title: Response[SDKSessionTokenSchema] HTTPValidationError: properties: detail: items: $ref: '#/components/schemas/ValidationError' type: array title: Detail type: object title: HTTPValidationError VideoSettings: properties: quality: $ref: '#/components/schemas/VideoQualityEnum' default: high encoding: $ref: '#/components/schemas/VideoEncodingEnum' default: H264 type: object title: VideoSettings description: Video encoding settings for session configuration AvatarPersonaSchema: properties: voice_id: title: Voice Id description: Voice ID. If empty use default avatar voice type: string format: uuid context_id: title: Context Id description: Context ID type: string format: uuid language: type: string title: Language description: Language default: en voice_settings: anyOf: - oneOf: - $ref: '#/components/schemas/ElevenLabsVoiceSettings' - $ref: '#/components/schemas/FishAudioVoiceSettings' - $ref: '#/components/schemas/CartesiaVoiceSettings' discriminator: propertyName: provider mapping: cartesia: '#/components/schemas/CartesiaVoiceSettings' elevenLabs: '#/components/schemas/ElevenLabsVoiceSettings' fish: '#/components/schemas/FishAudioVoiceSettings' - type: 'null' title: Voice Settings description: >- Additional voice settings (provider-specific, discriminated by 'provider' field) stt_config: anyOf: - $ref: '#/components/schemas/STTConfigSchema-Input' - type: 'null' description: ASR/STT configuration override llm_settings: anyOf: - $ref: '#/components/schemas/OpenAILLMSetting' - type: 'null' description: LLM selection (provider + model). None → the managed default model. type: object title: AvatarPersonaSchema VoiceAgentRef: properties: id: type: string format: uuid title: Id description: >- Stored voice_agent id. Its server-resolved agent_type determines which of the override fields below are accepted. language: title: Language description: >- Per-session language override, layered over the agent's stored language. Accepted only for a liveavatar_agent; rejected with a 400 for an elevenlabs_agent. type: string dynamic_variables: title: Dynamic Variables description: >- Per-session dynamic_variables override. Accepted only for a liveavatar_agent, where values must be strings (substituted as text into the context's ${var} placeholders; non-string values are rejected with a 400). Rejected with a 400 for an elevenlabs_agent. Mutually exclusive with the top-level dynamic_variables (400 if both set). At most 50 entries; keys <= 64 chars; string values <= 1000 chars. additionalProperties: anyOf: - type: string - type: integer - type: number - type: boolean type: object type: object required: - id title: VoiceAgentRef description: >- Reference to a stored voice_agent, plus optional per-session overrides. The caller passes only the agent ``id``; the server resolves the stored agent's ``agent_type`` and decides how the session is built. The override fields below are NOT universal — which ones are accepted depends on the resolved type, and an override that doesn't apply is rejected (400) rather than silently ignored: * ``liveavatar_agent`` (managed STT -> LLM -> TTS pipeline): ``language`` and ``dynamic_variables`` are honored, layered over the stored agent's config for this session only. * ``elevenlabs_agent`` (ElevenLabs-hosted): the stored agent is used as-is; passing ``language`` or ``dynamic_variables`` is rejected with a 400. Overrides are per-session-at-creation choices, not properties of the stored agent — the same agent is reused across sessions with different values each time, and the resolved result is what gets frozen into the session snapshot. Shared across the SDK and app token requests. ``dynamic_variables`` is typed as the bounded union for forward-compatibility (a single wire type across agent types); the server narrows or rejects it per the resolved ``agent_type`` as described on the field. SessionInteractivityTypeEnum: type: string enum: - CONVERSATIONAL - PUSH_TO_TALK title: SessionInteractivityTypeEnum MemoryRef: properties: prev_session_id: title: Prev Session Id description: >- Seed memory from a prior FULL-mode session. Reused if a session-memory already exists for that session; otherwise sync-bootstrapped. type: string format: uuid session_memory_id: title: Session Memory Id description: >- Attach an existing session-typed memory by id (must belong to caller's space). type: string format: uuid type: object title: MemoryRef description: |- Session-memory attach request. Either field may be set. If both are set, they must resolve to the same memory_id; mismatch -> 400. SDKSessionTokenSchema: properties: session_id: type: string format: uuid title: Session Id session_token: type: string title: Session Token type: object required: - session_id - session_token title: SDKSessionTokenSchema ValidationError: properties: loc: items: anyOf: - type: string - type: integer type: array title: Location msg: title: Message type: string type: type: string title: Error Type input: title: Input ctx: type: object title: Context type: object required: - loc - msg - type title: ValidationError VideoQualityEnum: type: string enum: - very_high - high - medium - low title: VideoQualityEnum VideoEncodingEnum: type: string enum: - VP8 - H264 title: VideoEncodingEnum ElevenLabsVoiceSettings: properties: provider: type: string const: elevenLabs title: Provider default: elevenLabs speed: type: number maximum: 1.2 minimum: 0.8 title: Speed default: 1 stability: type: number maximum: 1 minimum: 0 title: Stability default: 0.75 similarity_boost: type: number maximum: 1 minimum: 0 title: Similarity Boost default: 0.75 style: type: number maximum: 1 minimum: 0 title: Style default: 0 use_speaker_boost: type: boolean title: Use Speaker Boost default: true model: $ref: '#/components/schemas/ElevenLabsModelEnum' description: >- ElevenLabs TTS model. 'eleven_flash_v2_5' (default, fastest), 'eleven_multilingual_v2' (higher quality), 'eleven_v3', 'eleven_v3_conversational' and 'eleven_v4_turbo' (most expressive; these honor only 'stability' from the voice settings). default: eleven_flash_v2_5 apply_language_text_normalization: type: boolean title: Apply Language Text Normalization description: >- Apply ElevenLabs language-specific text normalization. Improves pronunciation for languages with complex orthography — currently only Japanese ('ja') is supported. Enabling this adds latency, so it is opt-in and only valid when the session language is set to 'ja'. default: false type: object title: ElevenLabsVoiceSettings description: ElevenLabs-specific voice settings for TTS configuration. FishAudioVoiceSettings: properties: provider: type: string const: fish title: Provider default: fish speed: type: number maximum: 1.2 minimum: 0.8 title: Speed default: 1 model: $ref: '#/components/schemas/FishAudioModelEnum' description: Fish Audio TTS model. default: s2.1-pro latency_mode: type: string enum: - normal - balanced - low title: Latency Mode description: >- Latency mode: 'normal' (best quality), 'balanced', or 'low' (lowest latency). default: balanced type: object title: FishAudioVoiceSettings description: Fish Audio-specific voice settings for TTS configuration. CartesiaVoiceSettings: properties: provider: type: string const: cartesia title: Provider default: cartesia speed: type: number maximum: 1.5 minimum: 0.6 title: Speed default: 1 model: $ref: '#/components/schemas/CartesiaModelEnum' description: Cartesia TTS model. default: sonic-3.6 type: object title: CartesiaVoiceSettings description: Cartesia-specific voice settings for TTS configuration. STTConfigSchema-Input: properties: provider: $ref: '#/components/schemas/STTProviderEnum' description: STT provider. Auto-selected if not specified. type: object required: - provider title: STTConfigSchema description: User-provided STT configuration for sessions. OpenAILLMSetting: properties: provider: type: string const: openai title: Provider default: openai model: $ref: '#/components/schemas/OpenAIModelEnum' default: gpt-5.4-nano type: object title: OpenAILLMSetting ElevenLabsModelEnum: type: string enum: - eleven_flash_v2_5 - eleven_multilingual_v2 - eleven_v3 - eleven_v3_conversational - eleven_v4_turbo title: ElevenLabsModelEnum FishAudioModelEnum: type: string enum: - s1 - s2 - s2-pro - s2.1-pro title: FishAudioModelEnum CartesiaModelEnum: type: string enum: - sonic-3.6 - sonic-3.5 - sonic-3 - sonic-2 - sonic-turbo title: CartesiaModelEnum STTProviderEnum: type: string enum: - deepgram - assembly_ai - gladia - elevenlabs title: STTProviderEnum OpenAIModelEnum: type: string enum: - gpt-4o - gpt-4o-mini - gpt-5-nano - gpt-5.4-nano title: OpenAIModelEnum description: Recommended OpenAI chat models selectable for the FULL pipeline. securitySchemes: APIKeyHeader: type: apiKey in: header name: X-API-KEY ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.