Lyria 3.5 vs MAI-Transcribe 2
Benchmark Performance
Available Benchmarks
Technical Differences
Side-by-Side Facts
| Field | Lyria 3.5 | MAI-Transcribe 2 |
|---|---|---|
| Developer | Google Deepmind | Microsoft |
| Family | Lyria | Mai Transcribe |
| Model | Lyria 3.5 | MAI-Transcribe 2 |
| Version | Lyria 3.5 | MAI-Transcribe 2 |
| Lifecycle | active | preview |
| Released | 2026-07-29 | 2026-09-03 |
| Knowledge cutoff | Unknown | Unknown |
| Input modalities | Text | Audio |
| Output modalities | Text, Audio | Text |
| Context window | 1,048,576 | Unknown |
| Total parameters | Unknown | Unknown |
| Active parameters | Unknown | Unknown |
| License | Unknown | Unknown |
| Open weights | No | No |
| API available | Yes | Yes |
| Self-hostable | No | Unknown |
| Provider access | Google AI (Standard), Google Gemini (Standard) | Microsoft Foundry (Public preview) |
| Capabilities | generation, image-to-music, lyrics, music-generation, synthid, vocals | diarization, multilingual, speaker-attribution, transcription, word-level-timestamps |
12 comparable fields · 10 material differences · Pair passes the primary-source comparison gate
Lyria 3.5 Capabilities
MAI-Transcribe 2 Capabilities
Internal Comparison Graph
Related Comparisons
| A | Pair | B | Context |
|---|---|---|---|
Lyria 3.5Google Deepmind | vs |  stable-audio-3-mediumStability AI | cross-developer peersaudio |
 MiniMax-Music3MiniMax | vs | Lyria 3.5Google Deepmind | cross-developer peersaudio |
VibeVoice-ASR-Streaming-1.5BMicrosoft | vs | MAI-Transcribe 2Microsoft | developer peersaudio, text |
VibeVoice-ASR-Streaming-7BMicrosoft | vs | MAI-Transcribe 2Microsoft | developer peersaudio, text |
| vs | MAI-Transcribe 2Microsoft | cross-developer peersaudio, text | |
| vs | MAI-Transcribe 2Microsoft | cross-developer peersaudio, text | |
 Muse Spark 1.1Meta | vs | MAI-Transcribe 2Microsoft | cross-developer peersaudio, text |
Muse Spark 1.2Meta | vs | MAI-Transcribe 2Microsoft | cross-developer peersaudio, text |
Lyria 3.5Google Deepmind | vs | InklingThinking Machines Lab | cross-developer peersaudio |
Lyria 3.5Google Deepmind | vs | Phi-4-multimodal-instructMicrosoft | cross-developer peersaudio |
Lyria 3.5Google Deepmind | vs | Muse Spark 1.1Meta | cross-developer peersaudio |
Lyria 3.5Google Deepmind | vs | Muse Spark 1.2Meta | cross-developer peersaudio |
Primary Evidence
Sources and Freshness
Questions
Lyria 3.5 vs MAI-Transcribe 2 FAQs
Is Lyria 3.5 or MAI-Transcribe 2 better for coding?+
This comparison does not currently contain a protocol-matched coding benchmark for both Lyria 3.5 and MAI-Transcribe 2, so Model Markets cannot name a coding leader from pricing, context size, or capability labels alone.
Which is cheaper, Lyria 3.5 or MAI-Transcribe 2?+
Neither model has a directly sourced input price in this comparison. Neither model has a directly sourced output price in this comparison.
Which has a larger context window, Lyria 3.5 or MAI-Transcribe 2?+
Neither model has a larger sourced context window in this comparison. Lyria 3.5 is 1,048,576 and MAI-Transcribe 2 is —.
Which performs better in benchmarks, Lyria 3.5 or MAI-Transcribe 2?+
There is no overall benchmark winner: At least two independently verified, protocol-matched benchmarks are required for an overall winner.
Can Lyria 3.5 or MAI-Transcribe 2 be self-hosted?+
Neither model is the only model in this pair currently marked as self-hostable. Lyria 3.5 is not marked open weight; MAI-Transcribe 2 is not marked open weight.
Can Lyria 3.5 and MAI-Transcribe 2 understand images?+
Lyria 3.5 is not documented with image input; MAI-Transcribe 2 is not documented with image input. This reflects supported input modalities, not vision quality.
Which can generate longer answers, Lyria 3.5 or MAI-Transcribe 2?+
Neither has a larger sourced maximum output. Lyria 3.5 is 65,536 and MAI-Transcribe 2 is —.
Do Lyria 3.5 and MAI-Transcribe 2 support reasoning and tool use?+
Lyria 3.5: none of these features are definitively sourced. MAI-Transcribe 2: none of these features are definitively sourced. Feature support does not establish relative quality.
Which is available from more inference providers, Lyria 3.5 or MAI-Transcribe 2?+
Lyria 3.5 has 2 sourced provider routes; MAI-Transcribe 2 has 1, so Lyria 3.5 has broader tracked availability.
Which offers better value, Lyria 3.5 or MAI-Transcribe 2?+
There is no universal value winner. Compare the input and output prices above with the matched benchmark result for your workload: cheaper tokens can be offset by different quality, token usage, latency, or provider availability.