Skip to content
Model infoModelsSnapshot: 15 Oct 2026

Alibaba Cloud / Fun-ASR

Fun-ASR

Transcribes long recordings with speaker labels, timestamps and configurable vocabulary.

availableClosed weightsSpeech-to-text
Official documentation (opens in a new tab)Reviewed 14 Oct 2026
24 of 31 specification fields collected
Audio durationUp to 12 hours per fileSee specification notes below
File size limit2 GB; one URL per taskSee specification notes below
Input billingSee ratesAlibaba Cloud Model Studio · Singapore
Output billingSee ratesPublished units and conditions below

Specifications

Provider-documented facts. “Not collected” means we have not verified that field yet, not that the feature is unsupported.

Context / input limit
Not collected
Maximum input tokens
Not collected
Maximum output
Not collected
Reasoning
Not collected
Streaming
Not supportedSource for Streaming (opens in a new tab)

Returns a task ID; poll for the finished result. Asynchronous processing is not streaming or discounted batch inference.

Knowledge cutoff
Not collected
Release date
2025-09-24Source for Release date (opens in a new tab)

Earliest reviewed Model Studio availability entry for this selector, not necessarily the private checkpoint training/release date.

Parameters
Not collected

The research repository reports 7.7B for its closed Fun-ASR evaluation model and 800M for public Nano models. Exact mapping to this hosted selector is not established.

Architecture
Not collected
Languages
View all 30 entries
  • Chinese
  • English
  • Japanese
  • Korean
  • Vietnamese
  • Thai
  • Indonesian
  • Malay
  • Filipino
  • Hindi
  • Arabic
  • French
  • German
  • Spanish
  • Portuguese
  • Russian
  • Italian
  • Dutch
  • Swedish
  • Danish
  • Finnish
  • Norwegian
  • Greek
  • Polish
  • Czech
  • Hungarian
  • Romanian
  • Bulgarian
  • Croatian
  • Slovak
Source for Languages (opens in a new tab)

Chinese dialect/accent coverage is model-specific. Stable Realtime and its February 2026 snapshot have different published language rosters.

Data retention
Standard API inference prompts and responses are retained up to 30 days, subject to legal/security exceptions; stateful files remain until deleted. Fun-ASR is absent from the published ZDR-supported model roster. Result URL expiry is not a general deletion guarantee.Source for Data retention (opens in a new tab)

Current service policy, not proof of deletion timing for past calls.

Usage metering
Input audio seconds; output text has no separate chargeSource for Usage metering (opens in a new tab)
Request mode
Asynchronous HTTP task with pollingSource for Request mode (opens in a new tab)
Audio duration limits
Up to 12 hours per fileSource for Audio duration limits (opens in a new tab)
Maximum audio file
2 GB; one URL per taskSource for Maximum audio file (opens in a new tab)

The file API accepts only one URL per request.

Word timestamps
Sentence and word timestamps; permanently enabledSource for Word timestamps (opens in a new tab)

Intermediate Flash timestamps can change until sentence completion.

Speaker diarization
SupportedSource for Speaker diarization (opens in a new tab)

Mono audio only; keep recordings within two hours when diarization is enabled.

Audio channels
Selectable tracks, with each requested track billed separatelySource for Audio channels (opens in a new tab)

Multi-channel transcription is distinct from speaker diarization; diarization is mono-only.

Result and task expiry
Task query and result download URL available for 24 hoursSource for Result and task expiry (opens in a new tab)
Custom vocabulary
Precompiled hotword lists; no instant weighted hotwordsSource for Custom vocabulary (opens in a new tab)

Main Fun-ASR permits up to 2,000 words per list; account limit is 10 lists. Instant weighted hotwords are limited to Qwen-Audio families.

Supplied context limit
API exposes context: up to 5 user and 5 assistant messages; ≤400 characters per turnSource for Supplied context limit (opens in a new tab)

The file API includes context, but the general context-enhancement roster omits Fun-ASR. The SDK does not yet support this field. Route/account availability requires confirmation.

Model revision
The current card identifies fun-asr with the 2025-11-07 snapshotSource for Model revision (opens in a new tab)

The named dated endpoint is separately retired; the undated selector is a recommended replacement. Endpoint retirement does not establish retirement of the shared model weights.

Access & pricing

Prices belong to an endpoint and billing tier. A provider listing does not guarantee access for every account or region.

Alibaba Cloud Model Studio · Singapore

Provider source (opens in a new tab)
API model ID
fun-asr
Access
POST https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1/services/audio/asr/transcription; X-DashScope-Async: enable; poll for results. Include parameters even when empty on the workspace domain.
Regions
Singapore. Deployment scope is not a country-only residency guarantee.
Endpoint availability
Listed in current Model Studio pricingSource for Endpoint availability (opens in a new tab)
Streaming
Not supportedSource for Streaming (opens in a new tab)

Returns a task ID; poll for the finished result. Asynchronous processing is not streaming or discounted batch inference.

Published request quota
600 requests/minuteSource for Published request quota (opens in a new tab)
Included free allowance
36,000 audio seconds; 90-day eligibility windowSource for Included free allowance (opens in a new tab)

From the latest of activation, release or access approval. It is not recurring free inference.

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 14 Oct 2026
CategoryUsageRateUnit
Speech recognitionInput audio1$0.000035second of input audio

Rate conditions

  1. Original regional list rate; transcription output has no separate charge. File transcription bills each requested track separately. Free quotas and promotional rates are separate; no generic text-question estimate.
Pricing source (opens in a new tab)

Data & enterprise terms

Standard API inference prompts and responses are retained up to 30 days, subject to legal/security exceptions; stateful files remain until deleted. Fun-ASR is absent from the published ZDR-supported model roster. Result URL expiry is not a general deletion guarantee.

Read the service policy (opens in a new tab)

Alibaba Cloud Model Studio · China (Beijing)

Provider source (opens in a new tab)
API model ID
fun-asr
Access
POST https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/api/v1/services/audio/asr/transcription; X-DashScope-Async: enable; poll for results. Include parameters even when empty on the workspace domain.
Regions
China (Beijing). Deployment scope is not a country-only residency guarantee.
Endpoint availability
Listed in current Model Studio pricingSource for Endpoint availability (opens in a new tab)
Streaming
Not supportedSource for Streaming (opens in a new tab)

Returns a task ID; poll for the finished result. Asynchronous processing is not streaming or discounted batch inference.

Published request quota
600 requests/minuteSource for Published request quota (opens in a new tab)
Included free allowance
No free quotaSource for Included free allowance (opens in a new tab)

Original international-account USD tariff; domestic-account promotions are separate.

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 14 Oct 2026
CategoryUsageRateUnit
Speech recognitionInput audio1$0.000032second of input audio

Rate conditions

  1. Original regional list rate; transcription output has no separate charge. File transcription bills each requested track separately. Free quotas and promotional rates are separate; no generic text-question estimate.
Pricing source (opens in a new tab)

Data & enterprise terms

Standard API inference prompts and responses are retained up to 30 days, subject to legal/security exceptions; stateful files remain until deleted. Fun-ASR is absent from the published ZDR-supported model roster. Result URL expiry is not a general deletion guarantee.

Read the service policy (opens in a new tab)

Alibaba Cloud Model Studio · Singapore

Provider source (opens in a new tab)
API model ID
fun-asr-2025-11-07
Access
Retired selector; do not treat this as a current access option. The provider recommends fun-asr.
Regions
Singapore · historical access record
Endpoint availability
Retirement announced for 2026-10-10; actual change time may varySource for Endpoint availability (opens in a new tab)

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 14 Oct 2026
CategoryUsageRateUnit
Historical speech recognitionInput audio1$0.000035second of input audio

Rate conditions

  1. Historical documented rate for a retired selector, not a current purchasable offer. Original regional price; no currency or audio-to-token conversion.
Pricing source (opens in a new tab)

Data & enterprise terms

Standard API inference prompts and responses are retained up to 30 days, subject to legal/security exceptions; stateful files remain until deleted. Fun-ASR is absent from the published ZDR-supported model roster. Result URL expiry is not a general deletion guarantee.

Read the service policy (opens in a new tab)

Alibaba Cloud Model Studio · China (Beijing)

Provider source (opens in a new tab)
API model ID
fun-asr-2025-11-07
Access
Retired selector; do not treat this as a current access option. The provider recommends fun-asr.
Regions
China (Beijing) · historical access record
Endpoint availability
Retirement announced for 2026-10-10; actual change time may varySource for Endpoint availability (opens in a new tab)

Endpoint pricing has not been collected. Check the provider for applicable rates.

Data & enterprise terms

Standard API inference prompts and responses are retained up to 30 days, subject to legal/security exceptions; stateful files remain until deleted. Fun-ASR is absent from the published ZDR-supported model roster. Result URL expiry is not a general deletion guarantee.

Read the service policy (opens in a new tab)

Caching

Caching can reuse earlier input processing. Availability, lifetime and billing depend on the endpoint. A cache hit still generates a new response.

01 / INITIAL INPUTProcess & retain

The system processes eligible input and retains reusable state.

02 / LATER REQUESTCheck available state

Reuse depends on matching input, session state and the cache policy.

03 / CACHE HITReuse & generate

Reusing state can reduce repeated work. Any charges follow the billing rules below.

Alibaba Cloud Model Studio · Singapore

fun-asr-2025-11-07
Not verified
How it works
Not collected
Minimum reusable input
Not collected
Lifetime & refresh
Not collected
How to enable it
Not collected
Usage fields
Not collected
Billing rules
Not collected
Limits & data boundaries
Not collected

No verified cache rate for this endpoint. An empty rate is not a zero charge.

Caching reviewed 14 Oct 2026

Alibaba Cloud Model Studio · China (Beijing)

fun-asr-2025-11-07
Not verified
How it works
Not collected
Minimum reusable input
Not collected
Lifetime & refresh
Not collected
How to enable it
Not collected
Usage fields
Not collected
Billing rules
Not collected
Limits & data boundaries
Not collected

No verified cache rate for this endpoint. An empty rate is not a zero charge.

Caching reviewed 14 Oct 2026

What the measurements say

No benchmark observations have been added for this model. We do not turn a vendor description into a performance score.

Latency and throughput depend on the host, request size, reasoning settings and load. There is no measured speed or global usage figure in this record.

The record behind the model

Why it is included

The official Fun-ASR repository explicitly labels its evaluated 7.7B Fun-ASR model as not open-source, separately from public 800M Nano checkpoints. Hosted release cards and API selectors establish these service identities. Extending family-level closed-weight classification to each private revision is an inference; exact research-checkpoint mapping and parameter counts are not copied onto every hosted release.

Classification evidence (opens in a new tab)

Lifecycle dates

Singapore documented availability24 Sept 2025Source (opens in a new tab)
China (Beijing) documented availability21 Nov 2025Source (opens in a new tab)

Sources reviewed