All Models
C
CosyVoice 2
budgetby FunAudioLLM
Model ID:
FunAudioLLM/CosyVoice2-0.5BVoice synthesis, TTS. Available through the llmair OpenAI-compatible API at https://api.llmair.ai/v1.
Modality
audio
Price / 1M
$0.03 in
$0.10 out
Context
8K
Provider
FunAudioLLM
Quick Start
Call FunAudioLLM/CosyVoice2-0.5B via the OpenAI-compatible API.
curl
curl https://api.llmair.ai/v1/chat/completions \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "FunAudioLLM/CosyVoice2-0.5B",
"messages": [{"role": "user", "content": "Hello!"}]
}'Capabilities
×
Text Generation
Not supported
✓
Streaming
Supported
×
JSON Mode
Not supported
×
Function Calling
Not supported
×
System Prompts
Not supported
×
Vision Input
Not supported
×
Image Generation
Not supported
✓
Audio
Supported
×
Embeddings
Not supported
×
Reasoning
Not supported
Pricing
All prices per 1 million tokens. Pay only for what you use.
| Type | Price |
|---|---|
| Input (prompt) | $0.0300 / 1M |
| Output (completion) | $0.1000 / 1M |
✓ Best For
- ▸Voice synthesis, TTS
- ▸High-volume, cost-sensitive workloads
- ▸Speech recognition and synthesis
⚠ Limitations
- ▸No image input or output
- ▸Lower reasoning depth than flagship models
- ▸No real-time internet access
- ▸Knowledge has a training cutoff