ChatGPT, Claude, Gemini — one endpoint, one key. Over 70% cheaper than direct access. Subscribe for even lower rates.
One endpoint for GPT, Claude, and Gemini. Switch models with a single parameter.
Pay with cryptocurrency. Fast, private, no intermediaries.
Repeated contexts cached automatically. Up to 90% savings on system prompts.
We never store your prompts or completions. Only token counts for billing.
If a cluster goes down, we switch to another instantly. Zero configuration.
Works with OpenAI, Anthropic, Google SDKs. Just change the base URL.
Per 1M tokens. All models on all plans.
| Provider | Model | Input | Output | Cached | Context |
|---|---|---|---|---|---|
| OpenAI | gpt-5.2 | $0.60$1.75 | $4.50$14.00 | $0.06$0.17 | 400k |
| OpenAI | gpt-5.2-codex | $0.60$1.75 | $4.50$14.00 | $0.06$0.17 | 400k |
| OpenAI | gpt-5.3-codex | $0.60$1.75 | $4.50$14.00 | $0.06$0.17 | 400k |
| OpenAI | gpt-5.1-codex-mini | $0.08$0.25 | $0.65$2.00 | $0.01$0.03 | 400k |
| Anthropic | claude-4.6-sonnet | $1.00$3.00 | $5.00$15.00 | $0.10 / 1.25$0.30 / 3.75 | 200k |
| Anthropic | claude-4.6-opus | $1.75$5.00 | $8.50$25.00 | $0.17 / 2.15$0.50 / 6.25 | 200k |
gemini-3.1-pro | $0.70$2.00 | $4.00$12.00 | $0.07$0.20 | 1M | |
gemini-3-flash | $0.15$0.50 | $1.00$3.00 | $0.01$0.05 | 1M |
Pay per token or subscribe for maximum savings.
5 days of full access
For individual developers
For teams & production
Pay as you go
Optimized for cost, not speed. Latency may be higher than direct API calls. Best for async workloads, batch processing, and non-time-critical tasks.
Recommended setup for most apps: unified key + OpenAI-compatible endpoint.
https://api.void-api.com/v1/chat/completionsclaude*, gemini*, gpt*API key headers supported
Authorization: Bearer <key>, x-api-key, x-goog-api-key, query key / api_key.
from openai import OpenAI
client = OpenAI(
api_key="sk-va-unified-...",
base_url="https://api.void-api.com/v1"
)
resp = client.chat.completions.create(
model="claude-sonnet-4-6",
messages=[{"role": "user", "content": "Hello"}]
)Unified (recommended):
Base URL: https://api.void-api.com/v1
Request URL: https://api.void-api.com/v1/chat/completions
Anthropic (native):
Base URL: https://api.void-api.com
Request URL: https://api.void-api.com/v1/messages
OpenAI-compatible:
Base URL: https://api.void-api.com/v1
Request URL: https://api.void-api.com/v1/chat/completions
Google (native):
Base URL: https://api.void-api.com/google
Request URL: https://api.void-api.com/google/v1beta/models/{model}:generateContent
Key compatibility:
- sk-va-unified-... works on all routes
- sk-va-anthropic-... works on Anthropic + /v1
- sk-va-google-... works on Google + /v1
- sk-va-openai-... works on OpenAI + /v1OpenAI-compatible mode:
Provider: OpenAI Compatible
Base URL: https://api.void-api.com/v1
(Cline sends /chat/completions itself)
Key: sk-va-unified-...
Anthropic mode:
Provider: Anthropic
Base URL: https://api.void-api.com
(Cline sends /v1/messages itself)
Key: sk-va-anthropic-... (or unified)No. We optimize for cost, not latency. Response times may vary and can be slower than direct API calls — sometimes significantly during high load. If you need consistent sub-1s latency, use official APIs directly.
No. We never log or store your prompts, completions, or any request content. We only track token counts for billing purposes. Your data passes through but is never persisted.
Two options: Pay-as-you-go — no subscription, pay only for tokens used at standard rates. Subscription plans — monthly fee unlocks lower per-token rates and free cached tokens. Check the pricing table above for exact rates.
Yes, but with caveats. We provide no SLA or uptime guarantees. Service may experience downtime or degraded performance without notice. Best suited for cost-sensitive workloads where occasional delays are acceptable — not for latency-critical or mission-critical systems.
Get your API key in 30 seconds. Start with pay-as-you-go, upgrade when needed.
Create account