Managed inference APIs for chat, reasoning, vision, embeddings, reranking, and speech — paired with on-demand Cloud GPU & CPU compute, editor tooling, and transparent usage billing in Algerian Dinar (DZD).
/v1/chat/completionsimport os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["DEVUP_API_KEY"],
base_url="https://api.devupai.com/v1",
)
res = client.chat.completions.create(
model="deepseek-ai/DeepSeek-V4-Pro",
messages=[
{"role": "user", "content": "Hello from Algeria!"}
],
)
print(res.choices[0].message.content)Onboarding Path
Works with existing OpenAI SDK code or a single cURL command
Core Documentation
Multi-turn completions with SSE token streaming, function calling, JSON Schema enforcement, and configurable reasoning effort.
Modern Responses API workflows, document & image understanding, visual generation, video synthesis, and prompt caching.
High-dimensional vector embeddings, cross-encoder reranking for precision RAG, speech recognition, and text-to-speech.
Dedicated NVIDIA GPU and multi-core CPU instances with SSH, JupyterLab, persistent network volumes, and per-second billing.
Official coding assistant extension for VS Code, Cursor, and Antigravity IDE with repository context and inline actions.
Drop-in setup guides for Python, Node.js, Go, Vercel AI SDK, LangChain, LlamaIndex, Claude Code, Cursor, and 30+ tools.
Architecture
Inspect context windows, pricing, supported modalities, and reasoning levels programmatically via GET /v1/models.
Keep existing OpenAI client code with a single base_url change, or install the typed devupai SDK for native ergonomics.
Transparent pay-as-you-go token and compute accounting in Algerian Dinar (DZD), without foreign-currency card friction.
Machine-readable error codes, explicit rate-limit headers, service tiers, and signed webhook delivery for reliable systems.
Copy-paste blueprints with verified request payloads and expected terminal outputs.