Introduction
Models
Public models available through the platform catalog.
The table below lists the models currently published in this deployment's catalog. Per-organization allow-lists may narrow what your API key can call; always query GET /v1/models for the authoritative list for your key.
| Model | Type | Context window | Vision | Structured output | Streaming |
|---|---|---|---|---|---|
| myt/deepseek-v4-flash | chat | 1M | No | Yes | Yes |
| myt/gemma-4-26b-a4b-it | chat | 256k | Yes | Yes | Yes |
| myt/glm5.2 | chat | 131k | No | Yes | Yes |
| myt/myt_asr | audio_transcription | 4k | — | — | Yes |
| myt/myt_tts | audio_speech | 4k | — | — | Yes |
| myt/nemotron-content-safety | chat | 128k | No | Yes | Yes |
| myt/qwen3.6-35b-a3b | chat | 131k | Yes | Yes | Yes |
Choosing a model
- Prefer smaller models when latency matters and quality requirements are moderate
- Use models with vision only when you pass image inputs
- Confirm structured output support before relying on constrained JSON mode