seed_audio_generate
Generate full-scene audio with Seed Speech (text prompt, optional voice-cloning audio or image reference).
Returns a durable artifact (provider URLs expire after 2 hours). variations (up to 5) run in parallel and return a summary. Run through ark_job_submit.
| Requirement | Value |
|---|---|
| Registration | Requires BYTEPLUS_SEED_SPEECH_API_KEY. |
| Execution | Background required |
| Required permission | seed:audio:generate |
This site lists the full configured capability surface, not the tools enabled on your particular server. Use model discovery and job discovery to check your connection. Credentials are server configuration, never tool arguments.
Supported Models
Section titled “Supported Models”These are supported families and built-in model or service IDs for this site revision, not provider account access or a guarantee that every row is enabled on your server. Check model discovery for configured bindings and service availability.
| Model | Model or service ID | Availability | Capabilities |
|---|---|---|---|
| Seed Audio 1.0 | seed-audio-1.0 |
BYTEPLUS_SEED_SPEECH_API_KEY required. |
Full-scene speech, music and sound generation; text, audio references or image reference; image and audio references are mutually exclusive; 1-5 independent variations. |
This tool has no per-call model argument. The Seed Speech adapter uses the fixed model below; this is not a ModelArk model binding.
Ask your connected AI client to use seed_audio_generate for the operation you want.
The client constructs the arguments and handles tool calls. Review its proposed
operation before approving anything billable. For a non-billable introduction,
start with discovery without generation.
Submit through ark_job_submit with tool_name and the complete target arguments, including its inner input object. Poll ark_job_get using the returned Ark job_id, respecting poll_after_ms. Inspect both terminal status and result.is_error; the original output is in result.structured_content. Do not call this tool in the foreground.
Exact MCP call (for developers)
The following is an MCP tool-call envelope, not a shell command. Example media URLs and identifiers must be replaced with your own accessible resources. Seedream and Seedance examples are non-billable dry runs; other provider operations may incur charges.
{ "name": "ark_job_submit", "arguments": { "input": { "tool_name": "seed_audio_generate", "arguments": { "input": { "text_prompt": "Gentle rain on a window. Max duration: 5 seconds" } } } }}Inputs
Section titled “Inputs”The MCP arguments object wraps these fields as {"input": {...}}.
| Field | Type | Required | Description and constraints |
|---|---|---|---|
audio_references |
array of AudioReference | No | Reference audio for voice cloning or scene control (max 3). Mutually exclusive with image_reference.; maxItems: 3 |
image_reference |
ReferenceImageInput or null | No | Reference image for visual-guided audio generation. Mutually exclusive with audio_references.; Default: null |
output |
AudioOutputOptions or null | No | Optional output format, sample rate, and rate controls.; Default: null |
output_path |
string or null | No | Optional local destination for the generated audio: a file, or a directory when it ends with ’/’ (files named <artifact_id>. |
overwrite |
boolean | No | Replace an existing file at output_path/output_dir when its content differs (identical content is always accepted).; Default: false |
persist |
boolean | No | Persist the audio as a durable MCP resource.; Default: true |
text_prompt |
string or null | No | Text describing the audio to generate (up to 3,000 characters). Required unless variation_prompts is set.; Default: null |
variation_prompts |
array of string or null | No | One prompt per variation, overriding text_prompt. Must have exactly variations entries.; Default: null |
variations |
integer | No | Parallel generations (1-5). Audio has no seed, so results differ by sampling; failures are reported per variation.; minimum: 1; maximum: 5; Default: 1 |
watermark |
AudioWatermarkOptions or null | No | Optional AIGC watermark settings.; Default: null |
Outputs
Section titled “Outputs”| Field | Type | Required | Description and constraints |
|---|---|---|---|
artifact |
ArtifactRef or null | No | Durable artifact reference for the generated audio; null when variations > 1.; Default: null |
billing_duration_seconds |
number or null | No | Billing-relevant audio duration in seconds (may differ from actual duration); null when variations > 1.; Default: null |
duration_seconds |
number or null | No | Actual audio duration in seconds; null when variations > 1.; Default: null |
model |
“seed-audio-1.0” | No | Model used for generation.; Default: “seed-audio-1.0” |
provider |
“byteplus-seed-speech” | No | Provider that generated the audio.; Default: “byteplus-seed-speech” |
provider_log_id |
string or null | No | Provider-side log ID for troubleshooting.; Default: null |
request_id |
string or null | No | Client request ID used for this generation.; Default: null |
source_url |
string or null | No | Provider HTTP URL (2-hour expiry) for direct download.; Default: null |
subtitle |
Subtitle or null | No | Timestamped subtitle data, if requested and available.; Default: null |
summary |
VariationSummary or null | No | Per-variation artifacts and errors when variations > 1; null otherwise.; Default: null |
Nested Types
Section titled “Nested Types”Field types named above refer to these shared definitions. Optional/null fields, enum choices and bounds are retained in the complete schemas below. Schema validation is followed by runtime capability and security checks.
AudioOutputOptions
| Field | Type | Required | Description and constraints |
|---|---|---|---|
format |
“wav”, “mp3”, “pcm”, “ogg” or null | No | Output audio format. Defaults to wav if not specified.; Default: null |
loudness_rate |
integer or null | No | Volume/loudness adjustment. -50 (quietest) to 100 (loudest).; Default: null |
pitch_rate |
integer or null | No | Pitch adjustment in semitones. -12 to +12.; Default: null |
sample_rate |
8000, 16000, 24000, 32000, 44100, 48000 or null | No | Output sample rate in Hz. Defaults to provider setting if not specified.; Default: null |
speech_rate |
integer or null | No | Speech speed adjustment. -50 (slowest) to 100 (fastest).; Default: null |
subtitle |
boolean or null | No | Whether to return timestamped subtitles in the response.; Default: null |
subtitle_type |
“utterance”, “word” or null | No | Subtitle granularity: utterance-level or word-level timestamps.; Default: null |
AudioReference
| Field | Type | Required | Description and constraints |
|---|---|---|---|
data |
string or null | No | Base64-encoded reference audio. Required when kind is ‘base64’.; Default: null |
kind |
“speaker”, “url”, “base64” | Yes | Reference mode: speaker ID, URL, or Base64 data. |
mime_type |
string or null | No | MIME type of the reference audio.; Default: null |
speaker_id |
string or null | No | Predefined speaker ID. Required when kind is ‘speaker’.; Default: null |
url |
string or null | No | HTTPS URL of the reference audio, or ‘asset://<asset_id>’ for an Active private Audio asset. Required when kind is ‘url’.; Default: null |
AudioWatermarkOptions
| Field | Type | Required | Description and constraints |
|---|---|---|---|
enable |
boolean or null | No | Enable or disable AIGC audio watermarking.; Default: null |
metadata |
boolean or null | No | Whether to embed AIGC metadata in the audio file.; Default: null |
MediaSourceKind
No fields. Pass {}.
ReferenceImageInput
| Field | Type | Required | Description and constraints |
|---|---|---|---|
data |
string or null | No | Base64-encoded media data. Required when kind is ‘base64’.; Default: null |
kind |
MediaSourceKind | Yes | Whether the media is referenced by URL or Base64. |
mime_type |
string or null | No | MIME type of the media (e.g. image/png, audio/wav).; Default: null |
url |
string or null | No | HTTPS URL of the media, or ‘asset://<asset_id>’ for an Active private asset library asset (ark_asset_create). Required when kind is ‘url’.; Default: null |
SeedAudioGenerateInput
| Field | Type | Required | Description and constraints |
|---|---|---|---|
audio_references |
array of AudioReference | No | Reference audio for voice cloning or scene control (max 3). Mutually exclusive with image_reference.; maxItems: 3 |
image_reference |
ReferenceImageInput or null | No | Reference image for visual-guided audio generation. Mutually exclusive with audio_references.; Default: null |
output |
AudioOutputOptions or null | No | Optional output format, sample rate, and rate controls.; Default: null |
output_path |
string or null | No | Optional local destination for the generated audio: a file, or a directory when it ends with ’/’ (files named <artifact_id>. |
overwrite |
boolean | No | Replace an existing file at output_path/output_dir when its content differs (identical content is always accepted).; Default: false |
persist |
boolean | No | Persist the audio as a durable MCP resource.; Default: true |
text_prompt |
string or null | No | Text describing the audio to generate (up to 3,000 characters). Required unless variation_prompts is set.; Default: null |
variation_prompts |
array of string or null | No | One prompt per variation, overriding text_prompt. Must have exactly variations entries.; Default: null |
variations |
integer | No | Parallel generations (1-5). Audio has no seed, so results differ by sampling; failures are reported per variation.; minimum: 1; maximum: 5; Default: 1 |
watermark |
AudioWatermarkOptions or null | No | Optional AIGC watermark settings.; Default: null |
Complete JSON Schemas
Section titled “Complete JSON Schemas”Input schema, including the MCP wrapper
{ "$defs": { "AudioOutputOptions": { "description": "Optional output controls for Seed Audio generation.", "properties": { "format": { "anyOf": [ { "enum": [ "wav", "mp3", "pcm", "ogg" ], "type": "string" }, { "type": "null" } ], "default": null, "description": "Output audio format. Defaults to wav if not specified." }, "loudness_rate": { "anyOf": [ { "maximum": 100, "minimum": -50, "type": "integer" }, { "type": "null" } ], "default": null, "description": "Volume/loudness adjustment. -50 (quietest) to 100 (loudest)." }, "pitch_rate": { "anyOf": [ { "maximum": 12, "minimum": -12, "type": "integer" }, { "type": "null" } ], "default": null, "description": "Pitch adjustment in semitones. -12 to +12." }, "sample_rate": { "anyOf": [ { "enum": [ 8000, 16000, 24000, 32000, 44100, 48000 ], "type": "integer" }, { "type": "null" } ], "default": null, "description": "Output sample rate in Hz. Defaults to provider setting if not specified." }, "speech_rate": { "anyOf": [ { "maximum": 100, "minimum": -50, "type": "integer" }, { "type": "null" } ], "default": null, "description": "Speech speed adjustment. -50 (slowest) to 100 (fastest)." }, "subtitle": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": null, "description": "Whether to return timestamped subtitles in the response." }, "subtitle_type": { "anyOf": [ { "enum": [ "utterance", "word" ], "type": "string" }, { "type": "null" } ], "default": null, "description": "Subtitle granularity: utterance-level or word-level timestamps." } }, "type": "object" }, "AudioReference": { "description": "An audio reference for Seed Audio, using one of three modes.", "properties": { "data": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Base64-encoded reference audio. Required when kind is 'base64'." }, "kind": { "description": "Reference mode: speaker ID, URL, or Base64 data.", "enum": [ "speaker", "url", "base64" ], "type": "string" }, "mime_type": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "MIME type of the reference audio." }, "speaker_id": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Predefined speaker ID. Required when kind is 'speaker'." }, "url": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "HTTPS URL of the reference audio, or 'asset://<asset_id>' for an Active private Audio asset. Required when kind is 'url'." } }, "required": [ "kind" ], "type": "object" }, "AudioWatermarkOptions": { "description": "AIGC watermark controls for Seed Audio.", "properties": { "enable": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": null, "description": "Enable or disable AIGC audio watermarking." }, "metadata": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": null, "description": "Whether to embed AIGC metadata in the audio file." } }, "type": "object" }, "MediaSourceKind": { "description": "Whether a media source is a URL or inline Base64 data.", "enum": [ "url", "base64" ], "type": "string" }, "ReferenceImageInput": { "description": "A reference image by URL, Base64 data, or private asset (``asset://<asset_id>``).", "properties": { "data": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Base64-encoded media data. Required when kind is 'base64'." }, "kind": { "$ref": "#/$defs/MediaSourceKind", "description": "Whether the media is referenced by URL or Base64." }, "mime_type": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "MIME type of the media (e.g. image/png, audio/wav)." }, "url": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "HTTPS URL of the media, or 'asset://<asset_id>' for an Active private asset library asset (ark_asset_create). Required when kind is 'url'." } }, "required": [ "kind" ], "type": "object" }, "SeedAudioGenerateInput": { "description": "Input model for ``seed_audio_generate``.", "properties": { "audio_references": { "description": "Reference audio for voice cloning or scene control (max 3). Mutually exclusive with image_reference.", "items": { "$ref": "#/$defs/AudioReference" }, "maxItems": 3, "type": "array" }, "image_reference": { "anyOf": [ { "$ref": "#/$defs/ReferenceImageInput" }, { "type": "null" } ], "default": null, "description": "Reference image for visual-guided audio generation. Mutually exclusive with audio_references." }, "output": { "anyOf": [ { "$ref": "#/$defs/AudioOutputOptions" }, { "type": "null" } ], "default": null, "description": "Optional output format, sample rate, and rate controls." }, "output_path": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Optional local destination for the generated audio: a file, or a directory when it ends with '/' (files named <artifact_id>.<ext>). A directory is required when variations > 1. stdio only; absolute path inside an allowed output root (client MCP roots or OUTPUT_ROOTS). The durable artifact is always kept; the copy is reported in ArtifactRef.local_path." }, "overwrite": { "default": false, "description": "Replace an existing file at output_path/output_dir when its content differs (identical content is always accepted).", "type": "boolean" }, "persist": { "default": true, "description": "Persist the audio as a durable MCP resource.", "type": "boolean" }, "text_prompt": { "anyOf": [ { "maxLength": 3000, "minLength": 1, "type": "string" }, { "type": "null" } ], "default": null, "description": "Text describing the audio to generate (up to 3,000 characters). Required unless variation_prompts is set." }, "variation_prompts": { "anyOf": [ { "items": { "maxLength": 3000, "minLength": 1, "type": "string" }, "type": "array" }, { "type": "null" } ], "default": null, "description": "One prompt per variation, overriding text_prompt. Must have exactly `variations` entries." }, "variations": { "default": 1, "description": "Parallel generations (1-5). Audio has no seed, so results differ by sampling; failures are reported per variation.", "maximum": 5, "minimum": 1, "type": "integer" }, "watermark": { "anyOf": [ { "$ref": "#/$defs/AudioWatermarkOptions" }, { "type": "null" } ], "default": null, "description": "Optional AIGC watermark settings." } }, "type": "object" } }, "additionalProperties": false, "properties": { "input": { "$ref": "#/$defs/SeedAudioGenerateInput" } }, "required": [ "input" ], "type": "object"}Output schema, including nested definitions
{ "$defs": { "ArtifactPersistenceIssue": { "description": "Safe explanation for a provider output that could not be persisted.", "properties": { "artifact_limit_bytes": { "description": "Maximum durable artifact size in bytes.", "title": "Artifact Limit Bytes", "type": "integer" }, "code": { "description": "Failure category.", "enum": [ "untrusted_output_host", "output_too_large", "invalid_output_mime", "source_expired", "download_failed", "storage_failed" ], "title": "Code", "type": "string" }, "message": { "description": "Safe failure message (no credentials or URLs).", "title": "Message", "type": "string" }, "retryable": { "description": "Whether retrying persistence later may succeed.", "title": "Retryable", "type": "boolean" }, "source_url_expires_at": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "ISO-8601 expiry of the unpersisted provider URL (the artifact uri).", "title": "Source Url Expires At" } }, "required": [ "code", "message", "retryable", "artifact_limit_bytes" ], "title": "ArtifactPersistenceIssue", "type": "object" }, "ArtifactRef": { "description": "Stable reference to a persisted media artifact.", "properties": { "bytes": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Size in bytes.", "title": "Bytes" }, "created_at": { "description": "ISO-8601 timestamp of artifact creation.", "title": "Created At", "type": "string" }, "expires_at": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "ISO-8601 expiry of the stored artifact.", "title": "Expires At" }, "export_error": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Why the requested local copy was not written (the artifact is unaffected).", "title": "Export Error" }, "fallback_data": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Base64 output bytes, only with id 'inline-fallback'.", "title": "Fallback Data" }, "id": { "description": "Unique artifact identifier.", "title": "Id", "type": "string" }, "local_path": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Absolute path of the local copy written for output_path/output_dir.", "title": "Local Path" }, "media_type": { "$ref": "#/$defs/MediaType", "description": "image, audio, video or three_d." }, "mime_type": { "description": "MIME type, e.g. image/png.", "title": "Mime Type", "type": "string" }, "persistence_error": { "anyOf": [ { "$ref": "#/$defs/ArtifactPersistenceIssue" }, { "type": "null" } ], "default": null, "description": "Set when the output was billed but not stored. id is then 'provider-url' (uri is the temporary provider URL; re-persist it with seed_media_persist_url before source_expires_at) or 'inline-fallback' (bytes in fallback_data)." }, "sha256": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "SHA-256 hex digest.", "title": "Sha256" }, "source_expires_at": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "ISO-8601 expiry of the provider URL.", "title": "Source Expires At" }, "uri": { "description": "MCP resource URI (seed-media://artifacts/{id}).", "title": "Uri", "type": "string" } }, "required": [ "id", "uri", "media_type", "mime_type", "created_at" ], "title": "ArtifactRef", "type": "object" }, "MediaType": { "description": "Logical generated-media categories.", "enum": [ "image", "audio", "video", "three_d" ], "title": "MediaType", "type": "string" }, "Subtitle": { "description": "Timestamped subtitle data for Seed Audio output.", "properties": { "utterances": { "description": "Utterance-level subtitle entries with timestamps.", "items": { "$ref": "#/$defs/SubtitleUtterance" }, "title": "Utterances", "type": "array" }, "words": { "description": "Word-level subtitle entries with timestamps.", "items": { "$ref": "#/$defs/SubtitleWord" }, "title": "Words", "type": "array" } }, "title": "Subtitle", "type": "object" }, "SubtitleUtterance": { "additionalProperties": true, "description": "Utterance-level subtitle timing returned by Seed Audio.", "properties": { "end": { "anyOf": [ { "type": "number" }, { "type": "null" } ], "default": null, "description": "End time in seconds.", "title": "End" }, "start": { "anyOf": [ { "type": "number" }, { "type": "null" } ], "default": null, "description": "Start time in seconds.", "title": "Start" }, "text": { "default": "", "description": "Utterance text.", "title": "Text", "type": "string" } }, "title": "SubtitleUtterance", "type": "object" }, "SubtitleWord": { "additionalProperties": true, "description": "Word-level subtitle timing returned by Seed Audio.", "properties": { "end": { "anyOf": [ { "type": "number" }, { "type": "null" } ], "default": null, "description": "End time in seconds.", "title": "End" }, "start": { "anyOf": [ { "type": "number" }, { "type": "null" } ], "default": null, "description": "Start time in seconds.", "title": "Start" }, "text": { "default": "", "description": "Word text.", "title": "Text", "type": "string" } }, "title": "SubtitleWord", "type": "object" }, "VariationError": { "description": "Machine-readable failure for one variation.", "properties": { "ambiguous_completion": { "default": false, "description": "Whether the provider may have partially completed despite the error.", "title": "Ambiguous Completion", "type": "boolean" }, "code": { "description": "Error code (e.g. PROVIDER_ERROR, TIMEOUT, UNEXPECTED_ERROR).", "title": "Code", "type": "string" }, "message": { "default": "", "description": "Human-readable error message.", "title": "Message", "type": "string" }, "phase": { "anyOf": [ { "enum": [ "queued", "generating", "persisting" ], "type": "string" }, { "type": "null" } ], "default": null, "description": "Stage at failure: 'queued' (never started, safe to retry), 'generating' (may have completed), or 'persisting' (generated, storing failed).", "title": "Phase" }, "request_id": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Provider request ID, if available.", "title": "Request Id" }, "retryable": { "default": false, "description": "Whether the caller may retry this variation.", "title": "Retryable", "type": "boolean" } }, "required": [ "code" ], "title": "VariationError", "type": "object" }, "VariationResult": { "description": "Result of a single variation within a parallel generation.", "properties": { "artifact": { "anyOf": [ { "$ref": "#/$defs/ArtifactRef" }, { "type": "null" } ], "default": null, "description": "Generated artifact (None if failed)." }, "error": { "anyOf": [ { "$ref": "#/$defs/VariationError" }, { "type": "null" } ], "default": null, "description": "Error if this variation failed." }, "index": { "description": "0-based variation index.", "title": "Index", "type": "integer" }, "provider_log_id": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Provider-side log ID for troubleshooting.", "title": "Provider Log Id" }, "request_id": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Request ID for this variation.", "title": "Request Id" }, "seed": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Seed used (if applicable).", "title": "Seed" }, "task_id": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Provider task ID, when the variation created a task.", "title": "Task Id" } }, "required": [ "index" ], "title": "VariationResult", "type": "object" }, "VariationSummary": { "description": "Aggregate result of a parallel generation.", "properties": { "failed": { "description": "Variations that failed.", "title": "Failed", "type": "integer" }, "succeeded": { "description": "Variations that produced output.", "title": "Succeeded", "type": "integer" }, "total": { "description": "Total variations requested.", "title": "Total", "type": "integer" }, "variations": { "description": "Per-variation results (artifacts, errors, and metadata).", "items": { "$ref": "#/$defs/VariationResult" }, "title": "Variations", "type": "array" } }, "required": [ "total", "succeeded", "failed" ], "title": "VariationSummary", "type": "object" } }, "description": "Output model for ``seed_audio_generate``.", "properties": { "artifact": { "anyOf": [ { "$ref": "#/$defs/ArtifactRef" }, { "type": "null" } ], "default": null, "description": "Durable artifact reference for the generated audio; null when variations > 1." }, "billing_duration_seconds": { "anyOf": [ { "type": "number" }, { "type": "null" } ], "default": null, "description": "Billing-relevant audio duration in seconds (may differ from actual duration); null when variations > 1.", "title": "Billing Duration Seconds" }, "duration_seconds": { "anyOf": [ { "type": "number" }, { "type": "null" } ], "default": null, "description": "Actual audio duration in seconds; null when variations > 1.", "title": "Duration Seconds" }, "model": { "const": "seed-audio-1.0", "default": "seed-audio-1.0", "description": "Model used for generation.", "title": "Model", "type": "string" }, "provider": { "const": "byteplus-seed-speech", "default": "byteplus-seed-speech", "description": "Provider that generated the audio.", "title": "Provider", "type": "string" }, "provider_log_id": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Provider-side log ID for troubleshooting.", "title": "Provider Log Id" }, "request_id": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Client request ID used for this generation.", "title": "Request Id" }, "source_url": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Provider HTTP URL (2-hour expiry) for direct download.", "title": "Source Url" }, "subtitle": { "anyOf": [ { "$ref": "#/$defs/Subtitle" }, { "type": "null" } ], "default": null, "description": "Timestamped subtitle data, if requested and available." }, "summary": { "anyOf": [ { "$ref": "#/$defs/VariationSummary" }, { "type": "null" } ], "default": null, "description": "Per-variation artifacts and errors when variations > 1; null otherwise." } }, "title": "SeedAudioGenerateOutput", "type": "object"}Troubleshooting
Section titled “Troubleshooting”- Missing tool: check provider credentials and feature flags, then restart the connection.
- Invalid arguments: keep the correct wrapper and follow current model limits from discovery. A valid schema alone does not prove the chosen model supports the combination.
- A job completes with an error: inspect the original tool result, not only the outer job status. Never automatically resubmit potentially billable work.
- Local output writes from jobs require server-configured
OUTPUT_ROOTS; HTTP cannot write to a client’s local disk.
See configuration, background execution, and artifact delivery.