NexGate

Available Models

Browse NexGate model IDs, categories, pricing, context windows, and selection guidance.

Model catalog

NexGate exposes verified chat models through one OpenAI-compatible endpoint. The live catalog is database-backed and available from the app and API.

Note

The database controls production availability. The docs list the current approved catalog and fallback rows, but /api/models is the runtime source for enabled models.

Approved model catalog

Model IDDisplay nameProviderCategoryContextInput/1MOutput/1M
gpt-5.5GPT-5.5OpenAIfrontier400K$5.00$30.00
gpt-5GPT-5OpenAIfrontier400K$1.25$10.00
gpt-5.4GPT-5.4OpenAIfrontier400K$2.50$15.00
gpt-4oGPT-4oOpenAIfrontier128K$2.50$10.00
gpt-4.1GPT-4.1OpenAIfrontier1,047,576$2.00$8.00
o4-minio4 MiniOpenAIreasoning200K$1.10$4.40
gpt-5.4-miniGPT-5.4 MiniOpenAIstandard400K$0.40$1.60
gpt-4.1-miniGPT-4.1 MiniOpenAIstandard1,047,576$0.40$1.60
gpt-5.4-nanoGPT-5.4 NanoOpenAIefficient400K$0.10$0.40
gpt-4.1-nanoGPT-4.1 NanoOpenAIefficient1,047,576$0.10$0.40
deepseek-v4-proDeepSeek V4 ProDeepSeekfrontier128K$1.93$3.83
deepseek-v3.2-specialeDeepSeek V3.2 SpecialeDeepSeekfrontier128K$0.40$0.80
deepseek-v3.2DeepSeek V3.2DeepSeekstandard128K$0.28$0.42
deepseek-v4-flashDeepSeek V4 FlashDeepSeekefficient128K$0.14$0.28
kimi-k2.6Kimi K2.6Moonshot AIfrontier256K$0.95$4.00
kimi-k2.5Kimi K2.5Moonshot AIstandard256K$0.60$2.50
grok-4-3Grok 4.3xAIfrontier128K$1.25$2.50
grok-4-20-reasoningGrok 4 20 ReasoningxAIreasoning128K$1.25$2.50
grok-4-1-fast-reasoningGrok 4.1 Fast ReasoningxAIreasoning128K$0.20$0.50
llama-3.3-70b-instructLlama 3.3 70B InstructMetastandard128K$0.13$0.40
llama-4-maverickLlama 4 MaverickMetastandard512K$0.18$0.70
llama-4-scoutLlama 4 ScoutMetaefficient512K$0.10$0.35

Categories

Frontier and premium

Use these for harder reasoning, long-context work, and tasks where quality matters more than price.

Examples: gpt-5.5, gpt-5, gpt-5.4, gpt-4.1, gpt-4o, deepseek-v4-pro, kimi-k2.6, grok-4-3.

Reasoning

Use these when the task benefits from explicit planning or deeper problem solving.

Examples: o4-mini, grok-4-20-reasoning, grok-4-1-fast-reasoning.

Standard

Use these as defaults for production apps, agent workflows, and coding tasks.

Examples: gpt-5.5, gpt-5.4-mini, deepseek-v3.2, kimi-k2.5, llama-4-maverick, llama-3.3-70b-instruct.

Efficient

Use these for high-volume routing, extraction, classification, and cost-sensitive chat.

Examples: gpt-4.1-nano, gpt-5.4-nano, deepseek-v4-flash, llama-4-scout.

How to use a model

response = client.chat.completions.create(
    model="gpt-5.5",
    messages=[{"role": "user", "content": "Hello"}],
)
const response = await client.chat.completions.create({
  model: "gpt-5.5",
  messages: [{ role: "user", content: "Hello" }],
});
curl https://api.nexgate.app/v1/chat/completions \
  -H "Authorization: Bearer ng-your-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.5",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

Availability and errors

Model availability depends on enabled database rows, provider credentials, credit balance, and spend safety settings.

ErrorMeaning
400 invalid_request_errorUnknown or disabled model
402 insufficient_creditsBalance cannot cover the estimated request
429 rate_limit_errorHourly spend safety limit reached
502 provider_errorUpstream provider failed

Warning

Model prices can change when upstream provider costs change. Credits remain USD-denominated and each request is charged at the NexGate price configured at the time of use.

What's next

On this page