Models
Every AI model available through LLMAI, with API slugs, context sizes, and token pricing.
Model Reference
Use the slug from the table below as the model field in your request. All models are available on every account — no model-specific unlocking or tier requirements.
OpenAI
| Model | API Slug | Context Window | Input / 1M tokens | Output / 1M tokens |
|---|---|---|---|---|
| GPT-5.4 | gpt-5.4 | 200K | $0.875 | $5.250 |
| GPT-5.3 Codex | gpt-5.3-codex | 200K | $0.613 | $4.900 |
| GPT-5.2 | gpt-5.2 | 128K | $0.613 | $4.900 |
| Model | API Slug | Context Window | Input / 1M tokens | Output / 1M tokens |
|---|---|---|---|---|
| Gemini 3.1 Pro Preview | gemini-3.1-pro-preview | 1M | $0.700 | $4.200 |
| Gemini 3 Flash Preview | gemini-3-flash-preview | 1M | $0.175 | $1.050 |
Z.AI
| Model | API Slug | Context Window | Input / 1M tokens | Output / 1M tokens |
|---|---|---|---|---|
| GLM 5 Turbo | glm-5-turbo | 128K | $0.420 | $1.400 |
| GLM 5 | glm-5 | 128K | $0.350 | $1.120 |
DeepSeek
| Model | API Slug | Context Window | Input / 1M tokens | Output / 1M tokens |
|---|---|---|---|---|
| DeepSeek V3.2 | deepseek-v3.2 | 64K | $0.280 | $0.840 |
Making a Request
Set the model field to the exact slug shown above:
curl https://api.llmai.dev/v1/chat/completions \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-v3.2",
"messages": [{"role": "user", "content": "Review this Python snippet for bugs."}]
}'Slugs are case-sensitive. GPT-5.4 and gpt-5.4 are not equivalent — use the lowercase slug exactly as shown.
Fetching the Model List Programmatically
curl https://api.llmai.dev/v1/models \
-H "Authorization: Bearer YOUR_API_KEY"Returns a standard ListModelsResponse object containing all model IDs currently available on the platform.
Notes on Pricing
- All rates are quoted per one million tokens.
- Actual charges scale proportionally — a request using 2,000 tokens costs (2,000 ÷ 1,000,000) × rate.
- Input and output tokens are priced independently.
- Context window is the combined maximum for prompt and completion tokens in a single call.
- Model availability is subject to change; check the Models catalog for the current live list.