Direct OpenAI-compatible API access for GPT-5.6 Sol, Terra & Luna models. Dedicated compute capacity, 4 concurrent sessions, and no per-token metering anxiety.
Official configuration files and test commands for Windows, Linux, and WSL.
{
"OPENAI_API_KEY": "vc-live-key-your-key-here"
}Predictable flat-fee billing with zero token overage anxiety. Upgrade or cancel anytime via Stripe.
Ideal for active builders & technical leads
Simultaneous active connection slots
No Token Limit per request
For heavy agent runs & continuous AI development
High throughput parallel processing
No Token Limit per request
We strip away token counting and proxy degradation, offering clean, direct API pipes for coding agents.
No extra rerouting, silent model swapping, or quantized proxies. The model you request is the exact model responding.
Manage scoped credentials, restrict access by IP range, rotation policies, and export environment variables instantly.
Round-the-clock infrastructure. Any announced maintenance time window is automatically compensated to your plan duration.
Send long contexts, full file context trees, and large diffs without getting hit by arbitrary token cap errors.
Drop-in replacement for OpenAI SDKs, LangChain, LlamaIndex, Cursor, VS Code, and Codex CLI by altering only the base URL.
Test your exact codebase and IDE integration before committing to a paid package with an operator-reviewed fit test.
Estimate how much you save on ViveCodex compared to standard metered pay-as-you-go OpenAI token rates.
Fully compliant with the OpenAI REST specification. Use standard headers and endpoints.
Pass your API key as a Bearer Token in the HTTP Authorization header:
Authorization: Bearer vc-live-your-api-key-here
Base Endpoint: https://vivecodex.com/v1
POST https://vivecodex.com/v1/chat/completions
{
"model": "gpt-5.6-sol",
"messages": [
{"role": "system", "content": "You are an expert software engineer."},
{"role": "user", "content": "Implement a binary search tree in Rust"}
],
"temperature": 0.2
}Query GET https://vivecodex.com/v1/models to inspect available high-performance models:
gpt-5.6-sol — High-throughput flagship reasoning & coding model.terra — Ultra-fast model optimized for IDE inline autocomplete & diff generation.luna — High-precision multimodal & deep context analysis model.| Code | Status Name | Description / Resolution |
|---|---|---|
401 |
Unauthorized | Invalid API key or inactive 30-day package. Contact support on Discord. |
429 |
Rate Limit Exceeded | Concurreny limit reached (e.g. > 2 or > 4 active sessions). Retry with exponential backoff. |
500 |
Provider Fail | Transient upstream network retry. Automatically routed to secondary buffer. |
Works natively with Cursor, VS Code, Codex CLI, Claude CLI, Hermes, and OpenAI SDKs.
Set Custom OpenAI Base URL to https://vivecodex.com/v1 in Cursor Settings.
Configure config.json with apiBase: "https://vivecodex.com/v1".
Configure ~/.codex/config.toml and ~/.codex/auth.json with ViveCodex base URL.
Pass base_url="https://vivecodex.com/v1" into official OpenAI SDK client instances.
You receive direct 30-day API access credentials (API Key and Base URL https://vivecodex.com/v1) with dedicated concurrent session slots (2 or 4 sessions) and no per-token billing caps.
Yes. The LLM models are official provider models without any extra rerouting or silent downgrades. The model you request is the exact model you receive responses from.
Our API is accessible 24/7. In the rare event of an announced provider maintenance window, your plan duration is automatically extended to compensate for the downtime.
Yes! Free 1-hour fit tests are available upon request and capacity. Contact us on Discord or email support to test your setup risk-free.
Plan prices ($100 for Pro, $180 for Ultra) cover 30 days of compute access. Payment provider transaction processing costs are calculated separately at checkout.