Qwen3.8-LiveTranslate Flash Realtime

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

Context / input limit

53,248 tokensSource for Context / input limit (opens in a new tab)

Maximum input tokens

49,152 tokensSource for Maximum input tokens (opens in a new tab)

Maximum output

4,096 tokensSource for Maximum output (opens in a new tab)

Input modalities

Audio, ImageSource for Input modalities (opens in a new tab)

Output modalities

Text, AudioSource for Output modalities (opens in a new tab)

Reasoning

Not collected

Tool calling

Not supportedSource for Tool calling (opens in a new tab)

True means supported on at least one reviewed regional endpoint; inspect access details for restrictions.

Structured output

Not supportedSource for Structured output (opens in a new tab)

True means supported on at least one reviewed regional endpoint; inspect access details for restrictions.

Streaming

SupportedSource for Streaming (opens in a new tab)

Batch processing

Not supportedSource for Batch processing (opens in a new tab)

True means supported on at least one reviewed regional endpoint; inspect access details for restrictions.

Fine-tuning

Not supportedSource for Fine-tuning (opens in a new tab)

True means supported on at least one reviewed regional endpoint; inspect access details for restrictions.

Knowledge cutoff

Not collected

Release date

Not collected

Parameters

Not collected

Architecture

Hybrid-MoE Thinker–Talker with interleaved audio and textSource for Architecture (opens in a new tab)

Languages

View all 60 entries

• Chinese (zh)
• English (en)
• Arabic (ar)
• German (de)
• French (fr)
• Spanish (es)
• Portuguese (pt)
• Indonesian (id)
• Italian (it)
• Korean (ko)
• Russian (ru)
• Thai (th)
• Vietnamese (vi)
• Japanese (ja)
• Turkish (tr)
• Hindi (hi)
• Malay (ms)
• Dutch (nl)
• Urdu (ur)
• Norwegian Bokmål (nb)
• Swedish (sv)
• Danish (da)
• Hebrew (he)
• Finnish (fi)
• Polish (pl)
• Icelandic (is)
• Czech (cs)
• Filipino (fil)
• Persian (fa)
• Cantonese (yue)
• Greek (el)
• Afrikaans (af)
• Asturian (ast)
• Belarusian (be)
• Bulgarian (bg)
• Bengali (bn)
• Bosnian (bs)
• Catalan (ca)
• Cebuano (ceb)
• Estonian (et)
• Galician (gl)
• Gujarati (gu)
• Croatian (hr)
• Hungarian (hu)
• Javanese (jv)
• Kazakh (kk)
• Kannada (kn)
• Kyrgyz (ky)
• Latvian (lv)
• Macedonian (mk)
• Malayalam (ml)
• Marathi (mr)
• Punjabi (pa)
• Romanian (ro)
• Slovak (sk)
• Slovenian (sl)
• Swahili (sw)
• Tajik (tg)
• Azerbaijani (az)
• Ukrainian (uk)

Source for Languages (opens in a new tab)

Supported audio-input and translated-text languages. Only the separately listed 29 languages have speech output.

Context caching

Not supportedSource for Context caching (opens in a new tab)

True means supported on at least one reviewed regional endpoint; inspect access details for restrictions.

API aliases

qwen3.8-livetranslate-flash-realtimeSource for API aliases (opens in a new tab)

Data retention

Standard inference prompts and responses are retained up to 30 days, subject to legal/security exceptions; stateful conversation storage and files remain until deleted. These Omni/LiveTranslate releases are not named in the current ZDR-supported roster.Source for Data retention (opens in a new tab)

Search grounding

Not supportedSource for Search grounding (opens in a new tab)

True means supported on at least one reviewed regional endpoint; inspect access details for restrictions.

Prefix completion

Not supportedSource for Prefix completion (opens in a new tab)

True means supported on at least one reviewed regional endpoint; inspect access details for restrictions.

Speech output languages

View all 29 entries

• Chinese (zh)
• English (en)
• Arabic (ar)
• German (de)
• French (fr)
• Spanish (es)
• Portuguese (pt)
• Indonesian (id)
• Italian (it)
• Korean (ko)
• Russian (ru)
• Thai (th)
• Vietnamese (vi)
• Japanese (ja)
• Turkish (tr)
• Hindi (hi)
• Malay (ms)
• Dutch (nl)
• Urdu (ur)
• Norwegian Bokmål (nb)
• Swedish (sv)
• Danish (da)
• Hebrew (he)
• Finnish (fi)
• Polish (pl)
• Icelandic (is)
• Czech (cs)
• Filipino (fil)
• Persian (fa)

Source for Speech output languages (opens in a new tab)

Language coverage

60 input/text-output languages; 29 spoken-output languagesSource for Language coverage (opens in a new tab)

Request mode

Realtime API: WebSocket, WebRTC or AOQSource for Request mode (opens in a new tab)

Speaker diarization

SupportedSource for Speaker diarization (opens in a new tab)

speaker_detection is the default turn-detection mode; server_vad is an alternative.

Utterance endpointing

speaker_detection by default; silence threshold fixed at 0.5 in that modeSource for Utterance endpointing (opens in a new tab)

Source-language transcription

Always enabled; source-language ASR transcripts are freeSource for Source-language transcription (opens in a new tab)

Voice cloning

SupportedSource for Voice cloning (opens in a new tab)

Disabled by default. once or always clones live input and requires voice=default; never requires a previously created voice ID.

Voice controls

Live cloning once per session or before each response; fixed pre-cloned profiles also documentedSource for Voice controls (opens in a new tab)

Stock voice catalog

Tina is the default system voiceSource for Stock voice catalog (opens in a new tab)

Voice enrollment cost

Real-time cloning has no separate fee. The guide lists USD 0.01 for advance voice creation, but enrollment target compatibility needs confirmation.Source for Voice enrollment cost (opens in a new tab)

Image input limits

JPG/JPEG up to 1080p; raw file ≤500 KB; Base64 transport; at most 2 images/second; send audio before the first imageSource for Image input limits (opens in a new tab)

Custom vocabulary

Hot-word mappings from source terms to translated termsSource for Custom vocabulary (opens in a new tab)

Usage metering

7 tokens per input audio second; 12.5 tokens per output audio second; each 32×32 image-pixel block uses 0.5 tokens. Source ASR text is free; translated text is a separate published output billing item.Source for Usage metering (opens in a new tab)

Translation pipeline

Automatic source-language detection when omitted; target language defaults to EnglishSource for Translation pipeline (opens in a new tab)

Provider-reported latency

Provider reports average lagging of 2.3 secondsSource for Provider-reported latency (opens in a new tab)

Reported evaluation result, not a latency guarantee for every request.

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

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

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.

Why it is included

Inference from the official API-served release and exact Model Studio cards, checked against the reviewed official Qwen public-weight inventory. No corresponding weights are published for this named release in that inventory. Public Qwen2.5-Omni and Qwen3-Omni checkpoints are excluded; their architectures and parameter counts are not treated as the hosted model identity.

Lifecycle dates

Sources reviewed