OpenAI-compatible API — switch models instantly by changing a single field.
Every request consumes credits equal to tokens processed multiplied by the model's credit multiplier. The base rate is 1.0×, ranging from 0.5× for lightweight models up to 3.5× for large frontier models.
| Multiplier | Tier | Description |
|---|---|---|
0.5× |
Standard — Lightweight | Small, ultra-fast models ideal for simple tasks |
1.0× |
Frontier — Embedding | Lightweight embedding models |
1.5× |
Standard — Mid | Mid-size models with strong performance |
2.5× |
Frontier — Mid | Balanced frontier models for quality and cost |
3.5× |
Frontier — Large | Highest-performance, largest-scale models |
Example: A request consuming 1,000 tokens on a 1.5× model deducts 1,500 credits. Monitor exact consumption via the usage field in every response.
Models optimized for fast performance and efficient credit usage. Base rate is 1 credit per token adjusted by the model multiplier.
Frontier models provide access to some of the most powerful AI models available, with tens or hundreds of billions of parameters. Multipliers range from 2.5× to 3.5×.
Retrieve the full list of available models via the following OpenAI-compatible endpoint:
GET https://llmapi.resayil.io/v1/modelscurl https://llmapi.resayil.io/v1/models \
-H "Authorization: Bearer YOUR_API_KEY"{
"object": "list",
"data": [
{
"id": "llama3.2:3b",
"object": "model",
"created": 1700000000,
"owned_by": "llm-resayil"
},
{
"id": "qwen3.5:397b",
"object": "model",
"created": 1700000000,
"owned_by": "llm-resayil"
}
]
}
Note: The endpoint is also accessible at GET /api/v1/models. Both paths return the same list.
All models share a single OpenAI-compatible endpoint. Simply change the model field to switch models:
POST https://llmapi.resayil.io/v1/chat/completions{
"model": "mistral-small3.2:24b",
"messages": [
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Explain quantum computing in simple terms."}
],
"temperature": 0.7,
"top_p": 0.9,
"max_tokens": 500,
"stream": false
}
Every response includes a usage field showing exact token consumption:
"usage": {
"prompt_tokens": 15,
"completion_tokens": 142,
"total_tokens": 157
}All subscription tiers have immediate access to all 36 models with no restrictions. The only differentiator is your available credit balance.
We continuously update the model catalog to include the latest and most capable models. New models appear immediately in GET /v1/models results and are ready to use.
If a model is deprecated, at least 30 days notice is provided along with migration guidance. Notifications are sent via email and dashboard alerts.
Learn about the credit system and billing to understand costs.
Go to Billing & Credits →