Algeria’s First AI Provider — Sovereign Data, Local Pricing in DZD
Built for businesses and developers who need secure, compliant, and locally hosted AI solutions.
Explore models and documentation before you sign in. After deployment, your workspace shows the live API base and keys — nothing sensitive is hardcoded on the marketing site.
Sovereign data
Host AI workloads aligned with Algeria-first compliance and residency expectations.
DZD-aligned pricing
Predictable economics for startups, enterprises, and public-sector teams.
OpenAI-compatible
Same chat and streaming semantics you already ship — swap the base URL and key.
Framework-ready
LangChain, LlamaIndex, and LangGraph work by pointing SDKs at your AIGrid endpoint.
OpenAI-compatible API
Same `/v1/chat/completions` contract you already know — point any OpenAI SDK or HTTP client at your AIGrid API base and bearer token.
Data sovereignty
Algeria’s first AI provider: keep org data, keys, and usage policy under your control — dedicated credentials per deployment with clear audit trails.
Framework-ready
LangChain, LangGraph, and LlamaIndex integrate by swapping `base_url` / `api_base` — no proprietary SDK required.
Model choice
Reasoning, OCR-capable, and high-throughput text models — pick the right profile for assistants, agents, and batch jobs.
Built for serious workloads
Each card summarizes modality, reasoning profile, and throughput orientation. Open a model to see curated integration examples that mirror how the in-app "How to use" dialog works after deployment.
Model type
Showing 6 of 6 models
Open-weight · Text LLM
gpt-oss-120b
General agents
gpt-oss-120b
Large open stack for general reasoning, code, and assistants.
Throughput
1,000 tokens/sec
- Strong default for assistants and tool loops
- OpenAI-compatible `/v1/chat/completions`
- Pair with LangGraph for stateful agents
Qwen · Text LLM
Qwen3-30B-A3B-Thinking
Reasoning
Qwen3-30B-A3B-Thinking
Reasoning-forward 30B tier for planning and analysis.
Throughput
660 tokens/sec
- Extended thinking style outputs
- Ideal for LangGraph flows
- Tune system prompts for chain-of-thought depth
Google · Text LLM
google/gemma-4-31B
High-volume chat
google/gemma-4-31B
Fast 31B text generation for assistants and RAG answers.
Throughput
580 tokens/sec
- Great for high-volume chat
- Pairs with LangChain ChatOpenAI
- Stable for LlamaIndex completion nodes
Zhipu / Z.ai · OCR
zai-org/GLM-OCR
PDF OCR
zai-org/GLM-OCR
Document and image OCR with multimodal chat messages.
Throughput
1.96 pages PDF/sec
- Vision + OCR style prompts
- Great for ingestion pipelines
- Resize images for latency
DeepSeek · OCR
DeepSeek-OCR
Image OCR
deepseek-ocr
OCR-focused multimodal stack for documents.
Throughput
860 tokens/sec
- Multimodal messages for scans
- Works with guarded BFF patterns
- Batch pages for throughput
Alibaba · Embedding
Alibaba-NLP/gte-Qwen2-7B-instruct
Embeddings
Alibaba-NLP/gte-Qwen2-7B-instruct
Dense text encoder for semantic search and clustering.
Throughput
No throughput info
- Vector / similarity workflows
- LlamaIndex Settings helper
- Keep API id in sync with the model catalog
Documentation
Step-by-step explanations sit next to every code block: when to use each client, how streaming behaves on the wire, and how major orchestration frameworks map to the AIGrid OpenAI-compatible API.
Minimal clients
curl, Node, browser fetch, and Python requests — with guidance on where secrets should live.
Open sectionLangChain & LCEL
Chat models, composable prompt chains, parsers, and a retrieval-style pipeline on your AIGrid deployment.
Open sectionLangGraph agents
Explicit graphs for multi-step flows, plus ReAct-style tool loops backed by the same deployment.
Open sectionLlamaIndex
Global LLM settings, vector indexes, query engines, and multi-turn chat aligned with AIGrid.
Open section