BUILT FOR YOUR NEXT AI PROJECT

Frontier models.
One way in.

From GPT and Claude to DeepSeek and Kimi. Bring more models into your application with the APIs you already know.

Explore modelsQuickstart
BASE URLhttps://openai-apikey-proxy.pages.dev/v1

OpenAI & Anthropic SDK compatible · Access depends on your key

THE MODELS YOU KNOW. ONE CONNECTION.
OpenAIClaudeDeepSeekKimiGLMMiniMax
MODEL CATALOG / AUTO SYNC
Live catalogCatalog updated

Find your model. Build what’s next.

Explore the spotlight, search the full catalog, and copy an exact model ID for your client.

95Catalog models11Model families4API protocols

Automatically deduplicated and refreshed about every 60 seconds while this page is visible, and on return. Not instant updates. If a refresh fails, the last valid catalog stays available.

95 matching models

Spotlight is an editorial selection, not a release-date or performance ranking.
GPTSpotlight

GPT-6 Astra

gpt-6-astra
ChatMessagesResponses
Service configured
ClaudeSpotlight

Claude Fable 5.1

claude-fable-5-1
ChatMessagesResponses
Service configured
GPTSpotlight

GPT-6.1 Sol

gpt-6.1-sol
ChatMessagesResponses
Service configured
ClaudeSpotlight

Claude Opus 5.5

claude-opus-5-5
ChatMessagesResponses
Service configured
DeepSeekSpotlight

DeepSeek V4 Pro 0813

deepseek-v4-pro-0813
ChatMessagesResponses
Service configured
KimiSpotlight

Kimi K3

kimi-k3
ChatMessagesResponses
Service configured
GLMSpotlight

GLM 5.3

glm-5p3
ChatMessagesResponses
Service configured
GrokSpotlight

Grok 4.7

grok-4.7
ChatMessagesResponses
Service configured
Claude

Claude Fable 5

claude-fable-5
ChatMessagesResponses
Service configured
Claude

Claude Haiku 4.5

claude-haiku-4-5
ChatMessagesResponses
Service configured
Claude

Claude Haiku 5.5

claude-haiku-5-5
ChatMessagesResponses
Service configured
Claude

Claude Opus 4.5

claude-opus-4-5
ChatMessagesResponses
Service configured
Claude

Claude Opus 4.6

claude-opus-4-6
ChatMessagesResponses
Service configured
Claude

Claude Opus 4.6 1M

claude-opus-4-6-1m
ChatMessagesResponses
Service configured
Claude

Claude Opus 4.7

claude-opus-4-7
ChatMessagesResponses
Service configured
Claude

Claude Opus 4.7 Extended Cache

claude-opus-4-7-extended-cache
ChatMessagesResponses
Service configured
Claude

Claude Opus 4.8

claude-opus-4-8
ChatMessagesResponses
Service configured
Claude

Claude Opus 5

claude-opus-5
ChatMessagesResponses
Service configured
Claude

Claude Opus 5.5 Extended Cache

claude-opus-5-5-extended-cache
ChatMessagesResponses
Service configured
Claude

Claude Opus 5 Low

claude-opus-5-low
ChatMessagesResponses
Service configured
Claude

Claude Sonnet 4.5

claude-sonnet-4-5
ChatMessagesResponses
Service configured
Claude

Claude Sonnet 4.6

claude-sonnet-4-6
ChatMessagesResponses
Service configured
Claude

Claude Sonnet 4.6 1M

claude-sonnet-4-6-1m
ChatMessagesResponses
Service configured
Claude

Claude Sonnet 5

claude-sonnet-5
ChatMessagesResponses
Service configured
Claude

Claude Sonnet 5.5

claude-sonnet-5-5
ChatMessagesResponses
Service configured
DeepSeek

DeepSeek V4 Flash

deep-seek-v4-flash
ChatMessagesResponses
Service configured
DeepSeek

DeepSeek V4 Flash Baseten

deep-seek-v4-flash-baseten
ChatMessagesResponses
Service configured
DeepSeek

DeepSeek V4 Flash No Think

deep-seek-v4-flash-no-think
ChatMessagesResponses
Service configured
DeepSeek

DeepSeek V4 Flash Vision Exp

deep-seek-v4-flash-vision-exp
ChatMessagesResponses
Service configured
DeepSeek

DeepSeek V4 Pro

deep-seek-v4-pro
ChatMessagesResponses
Service configured
DeepSeek

DeepSeek V4 Pro Baseten

deep-seek-v4-pro-baseten
ChatMessagesResponses
Service configured
DeepSeek

DeepSeek V4.1 Flash

deep-seek-v4.1-flash
ChatMessagesResponses
Service configured
DeepSeek

DeepSeek V4.1 Flash Low

deep-seek-v4.1-flash-low
ChatMessagesResponses
Service configured
DeepSeek

DeepSeek V4.1 Flash Max

deep-seek-v4.1-flash-max
ChatMessagesResponses
Service configured
DeepSeek

DeepSeek V4.1 Flash No Think

deep-seek-v4.1-flash-no-think
ChatMessagesResponses
Service configured
DeepSeek

DeepSeek V4.1 Flash Novita

deep-seek-v4.1-flash-novita
ChatMessagesResponses
Service configured
DeepSeek

DeepSeek Flash

deepseek-flash
ChatMessagesResponses
Service configured
DeepSeek

DeepSeek V4 Flash

deepseek-v4-flash
ChatMessagesResponses
Service configured
DeepSeek

DeepSeek V4 Pro

deepseek-v4-pro
ChatMessagesResponses
Service configured
DeepSeek

DeepSeek V4 Pro 0813 Baseten

deepseek-v4-pro-0813-baseten
ChatMessagesResponses
Service configured
Embedding

Doubao Embedding Vision

doubao-embedding-vision
Embeddings
Service configured
Embedding

Gemini Embedding 2 Preview

gemini-embedding-2-preview
Embeddings
Service configured
Embedding

Text Embedding 3 Large

text-embedding-3-large
Embeddings
Service configured
Embedding

Text Embedding 3 Small

text-embedding-3-small
Embeddings
Service configured
GLM

GLM 5

glm-5
ChatMessages
Service configured
GLM

GLM 5.2

glm-5p2
ChatMessagesResponses
Service configured
GLM

GLM 5.3 Flash

glm-5p3-flash
ChatMessagesResponses
Service configured
GLM

GLM 5.3 Flash Baseten

glm-5p3-flash-baseten
ChatMessagesResponses
Service configured
GLM

GLM 5.3 Flash Low

glm-5p3-flash-low
ChatMessagesResponses
Service configured
GLM

GLM 5.3 Flash No Think

glm-5p3-flash-no-think
ChatMessagesResponses
Service configured
GPT

GPT-5

gpt-5
ChatMessagesResponses
Service configured
GPT

GPT-5 Mini

gpt-5-mini
ChatMessagesResponses
Service configured
GPT

GPT-5 Nano

gpt-5-nano
ChatMessagesResponses
Service configured
GPT

GPT-5.1

gpt-5.1
ChatMessagesResponses
Service configured
GPT

GPT-5.2

gpt-5.2
ChatMessagesResponses
Service configured
GPT

GPT-5.3 Codex

gpt-5.3-codex
ChatMessagesResponses
Service configured
GPT

GPT-5.4

gpt-5.4
ChatMessagesResponses
Service configured
GPT

GPT-5.4 Mini

gpt-5.4-mini
ChatMessagesResponses
Service configured
GPT

GPT-5.4 Nano

gpt-5.4-nano
ChatMessagesResponses
Service configured
GPT

GPT-5.5

gpt-5.5
ChatMessagesResponses
Service configured
GPT

GPT-5.6 Luna

gpt-5.6-luna
ChatMessagesResponses
Service configured
GPT

GPT-5.6 Luna Max

gpt-5.6-luna-max
ChatMessagesResponses
Service configured
GPT

GPT-5.6 Sol

gpt-5.6-sol
ChatMessagesResponses
Service configured
GPT

GPT-5.6 Terra

gpt-5.6-terra
ChatMessagesResponses
Service configured
GPT

GPT-6 Luna

gpt-6-luna
ChatMessagesResponses
Service configured
GPT

GPT-6 Luna Low

gpt-6-luna-low
ChatMessagesResponses
Service configured
GPT

GPT-6 Sol

gpt-6-sol
ChatMessagesResponses
Service configured
GPT

GPT-6.1 Terra

gpt-6.1-terra
ChatMessagesResponses
Service configured
Grok

Grok 4.5

grok-4.5
ChatMessagesResponses
Service configured
Grok

Grok 4.6

grok-4.6
ChatMessagesResponses
Service configured
Kimi

Kimi K2.5

kimi-k2.5
ChatMessages
Service configured
Kimi

Kimi K2.5

kimi-k2p5
ChatMessagesResponses
Service configured
Kimi

Kimi K2.6

kimi-k2p6
ChatMessagesResponses
Service configured
Kimi

Kimi K2.7 Code

kimi-k2p7-code
ChatMessagesResponses
Service configured
MiniMax

MiniMax M2.5

MiniMax-M2.5
ChatMessagesResponses
Service configured
MiniMax

MiniMax M2.5 Highspeed

MiniMax-M2.5-highspeed
ChatMessagesResponses
Service configured
MiniMax

MiniMax M2.7

MiniMax-M2.7
ChatMessagesResponses
Service configured
MiniMax

MiniMax M2.7 Highspeed

MiniMax-M2.7-highspeed
ChatMessagesResponses
Service configured
MiniMax

MiniMax M2.7

minimax-m2p7
ChatMessagesResponses
Service configured
MiniMax

MiniMax M3

minimax-m3
ChatMessagesResponses
Service configured
MiniMax

MiniMax M3

MiniMax-M3
ChatMessagesResponses
Service configured
MiniMax

MiniMax M3 RFT Slide V1 Qat

minimax-m3-rft-slide-v1-qat
ChatMessagesResponses
Service configured
MiniMax

MiniMax M3.1 Flash

MiniMax-M3.1-Flash
ChatMessagesResponses
Service configured
MiniMax

MiniMax M3.1 Flash

minimax-m3p1-flash
ChatMessagesResponses
Service configured
Other

Mimo V2.6 Flash

mimo-v2.6-flash
ChatMessagesResponses
Service configured
Other

Mimo V2.6 Flash No Think

mimo-v2.6-flash-no-think
ChatMessagesResponses
Service configured
Other

Mimo V2.6 Pro

mimo-v2.6-pro
ChatMessagesResponses
Service configured
Other

Muse Spark Open

muse-spark-open
ChatMessagesResponses
Service configured
Other

Nemotron 3 Ultra

nemotron-3-ultra
ChatMessagesResponses
Service configured
Other

Nemotron 3.5 Super VL

nemotron-3.5-super-vl
ChatMessagesResponses
Service configured
Qwen

MiniMax M3 RFT Slide Qwen38 Baseten

minimax-m3-rft-slide-qwen38-baseten
ChatMessagesResponses
Service configured
Qwen

Qwen3.5 Plus

qwen3.5-plus
ChatMessages
Service configured
Qwen

Qwen3.6 Plus

qwen3.6-plus
ChatMessages
Service configured
Qwen

Qwen3.7 Plus

qwen3.7-plus
ChatMessages
Service configured
Trinity

Trinity Large Thinking

trinity-large-thinking
ChatMessagesResponses
Service configured

The public catalog reflects service configuration, not your authorized model list or a guarantee of successful inference. Query GET /v1/models with your own API key. Protocol support, availability and quotas depend on actual requests.

DEVELOPER FIRST

Less integration. More creation.

Keep your workflow. Get started with familiar tools and standard APIs.

  1. 01

    Get your API key

    Contact the platform service administrator privately to obtain a key and confirm your authorized models and quotas. Keep the key on your server.

    View integration guide
  2. 02

    Choose a model and protocol

    Copy an exact model ID. OpenAI URLs include /v1; Anthropic SDK base URLs do not.

    Models
  3. 03

    Make your first request

    Start with an SDK example below, then use streaming or image inputs where the model supports them.

    Examples

BUILD FROM AN EXAMPLE

Familiar SDKs. More possibilities.

Your model ID is already in the example. Keys come from environment variables; this page does not send inference requests.

Selecting a catalog model updates Chat examples. For other protocols, select a model that supports that protocol.

Code preview only. No inference credits used here.
REQUEST EXAMPLE
curl https://openai-apikey-proxy.pages.dev/v1/chat/completions \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-6-astra","messages":[{"role":"user","content":"Introduce yourself in one sentence"}]}'
https://openai-apikey-proxy.pages.dev/v1SDK COMPATIBLE

YOUR TOOLS, CONNECTED

Fits the way you build.

Connection details for popular clients. Model capabilities and client versions may differ.

Cursor / Cline / Continue / Cherry Studio

API format
Chat Completions (OpenAI)
Base URL
https://openai-apikey-proxy.pages.dev/v1
Model
as returned by /v1/models

Claude Code / Anthropic SDK

API format
Anthropic Messages
ANTHROPIC_BASE_URL
https://openai-apikey-proxy.pages.dev
Model
as returned by /v1/models

no /v1

Codex CLI / Responses API clients

API format
Responses (OpenAI)
Base URL
https://openai-apikey-proxy.pages.dev/v1
Model
as returned by /v1/models

DeepSeek Harness

API format
openai-completions
Base URL
https://openai-apikey-proxy.pages.dev/v1
Model
as returned by /v1/models

Hand-added models are treated as text-only. To send images, add defaultInput: [text, image] under the provider in settings.yaml.

NextChat / LobeChat / Chatbox

API format
Chat Completions (OpenAI)
Base URL
https://openai-apikey-proxy.pages.dev
Model
as returned by /v1/models

these clients append /v1 automatically

Dify / FastGPT (OpenAI-API-compatible)

API format
Chat Completions (OpenAI)
Base URL
https://openai-apikey-proxy.pages.dev/v1
Model
as returned by /v1/models

THE IMPLEMENTATION DETAILS

Need the implementation details?

Endpoint paths, authentication, and common errors in one place.

FAQ
Expand protocol & endpoint reference

Which base URL to use

It depends on the API format your client speaks. Clients append the format's path to the base URL, so a format whose path already contains /v1 (Anthropic Messages) must not have /v1 in the base URL — otherwise you get a 404.

Which base URL to use
Client API formatPath appended by clientBase URL to enterResulting request URL
Chat Completions/chat/completionshttps://openai-apikey-proxy.pages.dev/v1https://openai-apikey-proxy.pages.dev/v1/chat/completions
Responses/responseshttps://openai-apikey-proxy.pages.dev/v1https://openai-apikey-proxy.pages.dev/v1/responses
Anthropic Messages/v1/messageshttps://openai-apikey-proxy.pages.devhttps://openai-apikey-proxy.pages.dev/v1/messages

If a tool only asks for a single URL without a format choice, use https://openai-apikey-proxy.pages.dev/v1 (Chat Completions — a common integration choice; verify support for your model).

Endpoints

Send Authorization: Bearer <API_KEY> on every request. Anthropic Messages also accepts x-api-key.

Endpoints
MethodPathDescription
GET/v1/modelsList models available to the current key
POST/v1/chat/completionsOpenAI Chat Completions (most common; streaming & images)
POST/v1/responsesOpenAI Responses API
POST/v1/messagesAnthropic Messages API (x-api-key supported)
POST/v1/embeddingsOpenAI Embeddings
POST/v1/images/generationsOpenAI Images

Authentication

Use Authorization: Bearer sk-xxx. Anthropic Messages also accepts x-api-key: sk-xxx; sending both is fine — the server reconciles them. Keys must not contain leading or trailing whitespace.

GOOD TO KNOW

FAQ

A few things to know before you connect.

401 invalid API Key?

Check the header is Authorization: Bearer sk-xxx and that the key is enabled with remaining quota.

403 model_not_allowed / model_not_found?

model_not_allowed: your user group is not authorised for that model. model_not_found: the name does not exist. Check GET /v1/models first.

429?

A service rate, concurrency or quota limit may have been reached. Check the error details and retry policy.

502?

The upstream model service is temporarily unavailable. Retry shortly or switch models.

Should the base URL include /v1?

Yes for OpenAI formats (Chat Completions / Responses). No for Anthropic Messages. A few clients (e.g. NextChat) append /v1 themselves — give those the bare domain.

Which API format should I pick?

Default to Chat Completions — a common integration choice; verify support for your model. Use Anthropic Messages for Claude Code and other Anthropic-ecosystem tools, Responses for Codex CLI.

Is vision / image input supported?

Yes. Pass images as image_url (public URL or base64 data URL) in messages, or as an image block in Anthropic format. Pick a vision-capable model.

Can I call it from a browser?

Yes, CORS is enabled — but never expose your API key in public front-end code.

Your next idea starts here.

Choose a model. Copy the endpoint. Start building.

View integration guide