Model catalog
Pick a model, then call it.
Browse Sprag's current model catalog: pricing, supported parameters, and a request you can copy for each one.
Filter by modality
Text · Image · Video in
Qwen3.5 35B A3B
Omni-Modal
qwen/qwen3.5-35b-a3b
Mixture-of-experts text and vision-language model for coding, agentic tasks, and multimodal reasoning.
Input
Output
Context window
262K tokens
Audio in
Rhythm
Speech to text
rhythm
Automatic speech recognition model for multilingual speech-to-text transcription. Recognizes and identifies 30 languages, plus 22 Chinese dialects.
Input
Output
Context window
32.8K tokens
Text · Audio · Image · Video in
SPEV Decisions
Decision model
spev
Evaluate predicate, choice and score questions over shared text, native audio, one inline image or visual video frames. Returns ordered typed answers, probabilities and token usage.
Input
Output
Context window
8.2K tokens
Text · Image · Audio · Video in
Symphony
Omni-Modal
symphony
Realtime multimodal serve on full Qwen3-Omni (thinker + talker + code2wav): text, image, audio, and video in; text and speech out, with realtime tool calling. Supports text in 119 languages, speech understanding in 19, and speech generation in 10.
Input
Output
Context window
65.5K tokens
Text · Audio in
Chorus
Speech synthesis
chorus
Text-to-speech model that speaks in a built-in voice, a voice cloned from reference audio, or a voice designed from a description.
Input
Output
Context window
4.1K tokens