MODEL ID runware:dia2@2b

coming-soon

Dia2 2B

November 19, 2025

Dia2 2B is a 2 billion parameter streaming text-to-speech model from Nari Labs designed for real-time conversational AI. It begins generating audio immediately from partial text input, supports multi-speaker dialogue via speaker tags, voice cloning from a few seconds of reference audio, and non-verbal cues like laughter, sighs, and coughs. Released under Apache 2.0 for commercial use.

API Options

Platform-level options for task execution and delivery.

taskType string required value: audioInference: Identifier for the type of task being performed

taskUUID string required UUID v4: UUID v4 identifier for tracking tasks and matching async responses. Must be unique per task.

outputType string default: URL: Audio output type.

Allowed values 3 values

outputFormat string default: MP3: File format for the generated audio.

Allowed values 3 values

webhookURL string URI

Specifies a webhook URL where JSON responses will be sent via HTTP POST when generation tasks complete. For batch requests with multiple results, each completed item triggers a separate webhook call as it becomes available.

Learn more 1 resource

Webhooks
PLATFORM

deliveryMethod string default: sync

Determines how the API delivers task results.

Allowed values 2 values

: Returns complete results directly in the API response.
: Returns an immediate acknowledgment with the task UUID. Poll for results using getResponse.

Learn more 1 resource

Task Polling
PLATFORM

uploadEndpoint string URI

Specifies a URL where the generated content will be automatically uploaded using the HTTP PUT method. The raw binary data of the media file is sent directly as the request body. For secure uploads to cloud storage, use presigned URLs that include temporary authentication credentials.

Common use cases:

Cloud storage: Upload directly to S3 buckets, Google Cloud Storage, or Azure Blob Storage using presigned URLs.
CDN integration: Upload to content delivery networks for immediate distribution.

// S3 presigned URL for secure upload
https://your-bucket.s3.amazonaws.com/generated/content.mp4?X-Amz-Signature=abc123&X-Amz-Expires=3600

// Google Cloud Storage presigned URL
https://storage.googleapis.com/your-bucket/content.jpg?X-Goog-Signature=xyz789

// Custom storage endpoint
https://storage.example.com/uploads/generated-image.jpg

The content data will be sent as the request body to the specified URL when generation is complete.

ttl integer min: 60: Time-to-live (TTL) in seconds for generated content. Only applies when outputType is 'URL'.

includeCost boolean default: false: Include task cost in the response.

numberResults integer min: 1 max: 20 default: 1: Number of results to generate. Each result uses a different seed, producing variations of the same parameters.

Generation Parameters

Core parameters for controlling the generated content.

model string required value: runware:dia2@2b

Identifier of the model to use for generation.

Learn more 3 resources

negativePrompt string min: 2 max: 3000

Prompt to guide what to exclude from generation. Ignored when guidance is disabled (CFGScale ≤ 1).

Learn more 1 resource

Text To Image: Prompts Guiding The Generation
GUIDE

seed integer min: 0 max: 2147483647: Random seed for reproducible generation. When not provided, a random seed is generated in the unsigned 32-bit range.

CFGScale float min: 1 max: 30 step: 0.01 default: 3

Guidance scale representing how closely the images will resemble the prompt.

Learn more 1 resource

Text To Image: Cfg Scale Balancing Creativity And Control
GUIDE

speech object required

Settings for speech generation.

Properties 1 property

speech » text text string required min: 2 max: 3000: Text to convert to speech.

Settings

Technical parameters to fine-tune the inference process. These must be nested inside the settings object.

settings » temperature temperature float min: 0 max: 5 default: 0.6: Sampling temperature for text token generation. Lower = more coherent, higher = more variety. 0 = greedy decoding.

settings » audioTemperature audioTemperature float min: 0 max: 5 default: 0.8: Sampling temperature for audio token generation. Lower = cleaner audio, higher = more expressive. 0 = greedy decoding.

settings » includePrefix includePrefix boolean default: false: Whether to include prefix audio in the final output or trim it.

settings » prefixAudioSpeaker1 prefixAudioSpeaker1 string: Voice reference audio for speaker 1 ([S1]). Provide a short sample to clone voice style. Internally transcribed with Whisper for alignment.

settings » prefixAudioSpeaker2 prefixAudioSpeaker2 string: Voice reference audio for speaker 2 ([S2]). Requires prefixAudioSpeaker1 to also be set.

settings » topK topK integer min: 1 max: 500 default: 50: Top-K sampling for both text and audio tokens. Limits selection to K most likely tokens per step.

taskType string required value: audioInference: Type of the task.

taskUUID string required UUID v4: UUID of the task.

audioUUID string required UUID v4: UUID of the output audio.

audioURL string URI: URL of the output audio.

audioBase64Data string: Base64-encoded audio data.

audioDataURI string URI: Data URI of the output audio.

seed integer: The seed used for generation. If none was provided, shows the randomly generated seed.

cost float: Task cost in USD. Present when includeCost is set to true in the request.