Available Models
Access 15+ AI models from leading providers through a single API. Pro tier unlocks all models.
🚀 Quick Start
Use model selectors for automatic routing, or specify exact model IDs for control:
curl https://api.llm.kiwi/v1/chat/completions \
-H "Authorization: Bearer YOUR_API_KEY" \
-d '{
"model": "deepseek-v3.1", // or "pro", "fast", "default"
"messages": [{"role": "user", "content": "Hello!"}]
}'Recommended Selectors
defaultAuto-select best available model. Great for general use.
freefastLowest latency option for real-time applications.
freeproHighest quality with advanced reasoning capabilities.
proChat & Text Generation Models
| Model ID | Provider | Description | Input | Tier |
|---|---|---|---|---|
default | LLM.kiwi | Auto-select best model | text | free |
fast | LLM.kiwi | Low latency optimized | text | free |
pro | LLM.kiwi | Highest quality routing | text | pro |
gpt-4.1-nano-2025-04-14 | OpenAI | GPT-4.1 Nano - Fast & efficient | textimage | pro |
gpt-5-mini | OpenAI | GPT-5 Mini - Advanced reasoning | textimage | pro |
deepseek-v3.1 | DeepSeek | Excellent for coding & reasoning | textimage | pro |
mistral-small-3.1-24b-instruct-2503 | Mistral | Mistral Small 24B | textimage | pro |
codestral-2405 | Mistral | Code generation specialist | text | pro |
codestral-2501 | Mistral | Latest code model | text | pro |
ministral-8b-2512 | Mistral | Compact & fast | textimage | pro |
meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo | Meta | Llama 3.1 8B Turbo | text | pro |
gemini-2.5-flash-lite | Ultra fast Gemini | textimage | free | |
gemini-search | Web-grounded responses | textimage | pro | |
glm-4.5-flash | Zhipu AI | Chinese/English bilingual | textimage | pro |
bidara | BIDARA | Biomimicry design assistant | text | free |
Image Generation
fluxHigh-quality image generationAudio Transcription
whisperSpeech-to-text transcriptionList Models API
GEThttps://api.llm.kiwi/v1/models