NexGate
API reference

Models

Public model catalog endpoint with pricing, provider, category, context window, and fallback behavior.

Endpoint

GET https://api.nexgate.app/models

Returns all enabled chat models. Each entry includes the public model ID accepted by /api/v1/chat/completions, display metadata, and user-facing token prices.

Tip

Use this endpoint to populate model pickers. The live app also exposes the same catalog at nexgate.app/models.

Authentication

This endpoint is public and does not require an API key or active session.

Cache behavior

Cache-Control: public, s-maxage=60, stale-while-revalidate=300

If the database query fails, NexGate returns the built-in fallback catalog and adds:

x-nexgate-model-source: fallback

Warning

Treat fallback results as a temporary availability path. Model availability depends on which models are enabled, your credit balance, and spend safety settings.

Response fields

model_idstring

Public model ID to pass as the model value in chat completions.

display_namestring

Human-readable model name.

providerstring

Public provider or routing family: openai, deepseek, moonshot, xai, or meta.

categorystring

Catalog grouping: frontier, premium, reasoning, standard, or efficient.

context_windownumber

Advertised context size in tokens.

sell_input_per_millionstring

USD input-token price per 1 million tokens.

sell_output_per_millionstring

USD output-token price per 1 million tokens.

use_case_hintstring

Short guidance for choosing the model.

speed_badgestring

UI hint for speed, context size, or model family.

Example

curl https://api.nexgate.app/models
type NexGateModel = {
  model_id: string;
  display_name: string;
  provider: string;
  category: string;
  context_window: number;
  sell_input_per_million: string;
  sell_output_per_million: string;
  use_case_hint: string;
  speed_badge: string;
};

const response = await fetch("https://api.nexgate.app/models");
const models = (await response.json()) as NexGateModel[];

Success response

[
  {
    "model_id": "gpt-4.1-mini",
    "display_name": "GPT-4.1 Mini",
    "provider": "openai",
    "category": "standard",
    "context_window": 1047576,
    "sell_input_per_million": "0.4000",
    "sell_output_per_million": "1.6000",
    "use_case_hint": "Reliable workhorse",
    "speed_badge": "1M ctx"
  }
]

Current fallback catalog

The production database is the source of truth. The fallback registry returns exactly these 23 approved model IDs when the database is unavailable:

Model IDProviderCategoryContextInput/1MOutput/1M
gpt-5.5openaifrontier400K$5.00$30.00
gpt-5openaifrontier400K$1.25$10.00
gpt-5.4openaifrontier400K$2.50$15.00
gpt-4oopenaifrontier128K$2.50$10.00
gpt-4.1openaifrontier1,047,576$2.00$8.00
o4-miniopenaireasoning200K$1.10$4.40
gpt-5.4-miniopenaistandard400K$0.40$1.60
gpt-4.1-miniopenaistandard1,047,576$0.40$1.60
gpt-5.4-nanoopenaiefficient400K$0.10$0.40
gpt-4.1-nanoopenaiefficient1,047,576$0.10$0.40
deepseek-v4-prodeepseekfrontier128K$1.93$3.83
deepseek-v3.2-specialedeepseekfrontier128K$0.40$0.80
deepseek-v3.2deepseekstandard128K$0.28$0.42
deepseek-v4-flashdeepseekefficient128K$0.14$0.28
kimi-k2.6moonshotfrontier256K$0.95$4.00
kimi-k2.5moonshotstandard256K$0.60$2.50
grok-4-3xaifrontier128K$1.25$2.50
grok-4-20-reasoningxaireasoning128K$1.25$2.50
grok-4-1-fast-reasoningxaireasoning128K$0.20$0.50
llama-3.3-70b-instructmetastandard128K$0.13$0.40
llama-4-maverickmetastandard512K$0.18$0.70
llama-4-scoutmetaefficient512K$0.10$0.35

On this page