Get immediate access to fully managed AI models delivered via secure APIs on Australia's most powerful sovereign GPU infrastructure.
Production-ready AI inference
What you can build on Sparc
Language, vision, audio, image generation, search, and reranking - all on one OpenAI-compatible API.
Conversational AI
Customer support bots, internal helpdesk assistants, document Q&A, multilingual chat
Agentic workflows
Multi-step automation, reasoning, planning and execution pipelines.
Voice and audio
Voice transcription, call centre compliance, podcast processing, voice-to-workflow pipelines.
Vision and document
Visual QA, scanned document processing, image understanding.
Search and RAG
Semantic search, knowledge base Q&A, product catalogue retrieval, compliance document lookup.
Image Generation
Product imagery, marketing creative, personalised visuals.
Platform features
Instant model access
Start using AI models instantly through our OpenAI-compatible APIs. No migration, no headaches — just plug in and go.
Secure and sovereign
Data stays in Australia. Built for organisations that require compliance and security.
Full operational visibility
Manage API keys, teams, projects, and track usage in real time from one platform.
Everything you need to ship inference at scale
-
Plug into our OpenAI-compatible API and start building immediately.
-
Test before you build. Explore every available model instantly - no code, no friction.
-
Control access for every user across teams and projects. Real-time usage metrics and live dashboards give you per-project consumption visibility at a glance.
-
Restrict API key usage by IP address for an additional layer of security and access control.
Model Catalogue
Curated range of models including LLMs, vision, audio, image generation, and embeddings.
All accessible through a single, OpenAI-compatible endpoint.
Your existing code works on day one.
All prices in AUD.
| Model | Type | Context | Pricing |
|---|---|---|---|
| Black Forest Labs: FLUX.1 Schnell black-forest-labs | Image | 77 | $0.022/img |
| Google: Gemma 3 27B Instruct google | Text | 128K |
In $0.120
Out $0.500
|
| Meta: Llama 3.1 8B Instruct meta | Text | 131K |
In $0.030
Out $0.070
|
| MiniMax: M2.5 minimax | Text | 197K |
In $0.400
Out $1.600
|
| NVIDIA: Cosmos Reason2 8B nvidia | Text | 262K |
In $0.500
Out $2.500
|
| NVIDIA: Llama 3.2 EmbedQA 1B v2 nvidia | 8K | $0.020/M | |
| NVIDIA: Llama 3.2 RerankQA 1B v2 nvidia | Reranker | 8K | $0.020/M |
| NVIDIA: Nemotron 3 Nano 30B-A3B nvidia | Text | 131K |
In $0.080
Out $0.350
|
| NVIDIA: Nemotron Nano 2 VL nvidia | Vision | 33K |
In $0.280
Out $0.850
|
| NVIDIA: Whisper Large V3 nvidia | Audio | 448 | $0.0007/min |
| OpenAI: GPT-OSS 120B openai | Text | 128K |
In $0.180
Out $0.800
|
| Whisper Large V3 Turbo openai | Audio | 448 | $0.0003/min |
ResetData, built for Australia’s AI Future
-
Built and operated entirely in Australia, aligned with national strategy.
-
ISO27001, SOC 2, PSPF, and SCEC certified, trusted by Tier 1 enterprises.
-
Liquid cooling with 10x density and 45% fewer emissions.
-
Rapid deployment of AI capacity, from proof-of-concept to full-scale rollout.