Chat & LLM API — every model, one endpoint
GPT-5 · Claude Opus 4.6 · Claude Sonnet 4.5 · Gemini 3 Pro · DeepSeek v3 · Mistral Large. Same /chat/completions request shape, streaming SSE support, up to 200k context.
What it's for
Common workflows.
✓
Chat product backends — agents, copilots, assistants
✓
Content generation — blog posts, product descriptions, emails
✓
Code tasks — GPT-5 Codex, Claude Sonnet for agentic coding
✓
Data extraction and transformation
✓
Translation and localization
✓
Research and analysis — long-context summarization
API call
Ship in 30 seconds.
curl https://aimarcusimage.eu/api/v1/chat/completions \
-H "Authorization: Bearer sk-aig-..." \
-H "Content-Type: application/json" \
-d '{
"model": "claude-sonnet-4-5",
"messages": [
{ "role": "user", "content": "Explain streaming SSE to a junior dev in 3 paragraphs." }
],
"stream": true
}'
Models
Pick the model that fits.
All billed from the same credit balance.
chat
$10.0100 / million tokens
claude-opus-4-5, Chat, Output
Anthropic
chat
$10.0100 / million tokens
Claude-Opus-4-6, chat, Output
Anthropic
chat
$7.8400 / million tokens
gpt-5.4, Chat, Output
OpenAI
chat
$7.8400 / million tokens
gpt-5.2-codex, Chat, Output
OpenAI
chat
$7.8400 / million tokens
gpt-5.4-codex, Chat, Output
OpenAI
chat
$7.8400 / million tokens
gpt-5.3-codex, Chat, Output
OpenAI
chat
$5.9850 / million tokens
Claude-Sonnet-4-6, chat, Output
Anthropic
chat
$5.9850 / call
claude-sonnet-4-5, Chat, Output
Anthropic
chat
$5.6000 / million tokens
gpt-5.1-codex, Chat, Output
OpenAI
chat
$5.6000 / million tokens
gpt-5-codex, Chat, Output
OpenAI
chat
$4.9000 / million tokens
Gemini 3 Pro, Chat, Output
Google
chat
$4.9000 / million
Gemini 3.1 Pro- openai, chat, output
Google
FAQ
Questions, answered.
Is it OpenAI-compatible?+
Yes. POST /api/v1/chat/completions with {model, messages, stream}. The response shape matches OpenAI exactly — drop-in replacement for most SDKs.
Does streaming work?+
Yes — SSE streaming for GPT, Claude and most OpenRouter-backed models. Just set stream=true in the payload.
What about long context?+
Gemini 3 Pro supports up to 2M tokens context. Claude Opus 4.6 supports 200k. See each model page for the exact limit.
Ready to start?
Sign up, get $10 free credits, ship your first chat generation in under a minute.