Hyphen Gateway API
One key, one base URL, real model names.
The Hyphen gateway is an OpenAI-compatible LLM proxy. A single Hyphen
key — sk-… — authenticates every endpoint below. Get one at
app.hyphen-solution.com (pick a plan or
credit pack, then Dashboard → API Keys).
Authentication
All requests use HTTP Bearer auth:
Code
Models
The gateway serves ten models — real provider model names, all from MiniMax:
| Model | Use for |
|---|---|
minimax-m3 | Flagship — newest M-series generation |
minimax-m2.7 | Balanced daily driver for most agent work |
minimax-m2.5 | Fast — cheap, high-volume calls |
minimax-m2.7-highspeed | Latency-tuned minimax-m2.7 (~2x price) |
minimax-m2.5-highspeed | Latency-tuned minimax-m2.5 (~2x price) |
minimax-m2.1-highspeed | Latency-tuned minimax-m2.1 (~2x price) |
minimax-m2.1 | Previous-generation midrange — price-sensitive |
minimax-m2 | Oldest M-series generation — for legacy configs |
minimax-m2-her | Role-play / dialogue chat (66k context) |
minimax-text-01 | Long-context text |
These are the exact IDs the gateway accepts — there are no invented
aliases, and requesting any other ID returns an error. The catalog can
change with notice as models are added or retired. During a provider
incident a request may be served by a smaller model of the same
family; the model field in the response always reports the model
that actually served the request.
Budgets & caps
Every subscription key carries a monthly budget that resets each cycle.
When the budget is exhausted the gateway returns HTTP 429 with the
reset date — there is no surprise overage. Prepaid credit funds a separate
pay-as-you-go key with a non-resetting balance you can switch to. See the
429 responses documented on each endpoint.