1. Quickstart
Use your existing OpenAI SDK. Only change base URL tohttps://gateway.arkagentic.com/v1and pass your ArkAgentic API key.
cURL Β· bash
curl https://gateway.arkagentic.com/v1/chat/completions \
-H "Authorization: Bearer ***" \
-H "Content-Type: application/json" \
-d '{
"model": "ark-gpt-4o",
"messages": [{"role":"user","content":"Hello from ArkAgentic"}],
"temperature": 0.2,
"stream": false
}'2. Model Routing Directory
Pass standard ArkAgentic model IDs in model. Gateway can map aliases and upstream overrides internally while your client integration stays stable.
| Model ID | Category | Alias / Route Target | Notes |
|---|---|---|---|
| ark-claude-3-5-sonnet | Anthropic | claude-3-5-sonnet | General production chat workload |
| ark-claude-opus-5 | Anthropic | claude-opus-5 | General production chat workload |
| ark-claude-sonnet-5 | Anthropic | claude-sonnet-5 | General production chat workload |
| ark-cohere-embed-v3 | OpenAI | Cohere-embed-v3-multilingual | Vector embedding and semantic indexing |
| ark-cohere-rerank-v4-fast | OpenAI | Cohere-rerank-v4.0-fast | Retrieval reranking and relevance scoring |
| ark-cohere-rerank-v4-pro | OpenAI | Cohere-rerank-v4.0-pro | Retrieval reranking and relevance scoring |
| ark-deepseek-r1 | DeepSeek | deepseek-reasoner | Deep reasoning and long-chain analysis tasks |
| ark-deepseek-v4-flash | DeepSeek | DeepSeek-V4-Flash | General production chat workload |
| ark-deepseek-v4-pro | DeepSeek | DeepSeek-V4-Pro | General production chat workload |
| ark-gemini-1.5-pro | OpenAI | gemini-1.5-pro | General production chat workload |
For detailed token pricing and compute estimator, visitPricing β.
View all 19 models in Model Center β3. Advanced Gateway Features
Failover & Auto-Retry
If an upstream node is degraded or unavailable, gateway automatically reroutes and retries to healthy candidates with no client-side route change.
Streaming SSE
Standard OpenAI-compatible streaming is supported. Setstream: trueand consume Server-Sent Events in your existing client.
Streaming Request
{
"model": "ark-gpt-4o",
"messages": [{"role":"user","content":"stream this response"}],
"stream": true
}HTTP Status Codes & Error Handling
| Code | Meaning | Suggested Action |
|---|---|---|
| 400 | Invalid request payload | Validate JSON shape and model ID. |
| 401 | Unauthorized | Check Bearer token and key scope. |
| 429 | Rate limited | Apply exponential backoff and retry. |
| 5xx | Transient upstream/gateway failure | Retry with jitter; gateway may auto-failover. |