Background texture

一个 API, 所有模型

通过一个兼容 OpenAI 的端点访问 OpenAI、Anthropic、Google 等模型。推理成本零加价。

对话补全
~/modelmax $ curl https://api.modelmax.io/v1/chat/completions \ -H "Content-Type: application/json" \ -H "Authorization: Bearer sk-..." \ -d '{
    "model": "gpt-4o",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'
视频生成
~/modelmax $ curl -X POST https://api.modelmax.io/v1/queue/veo-2.0-high \ -H "Content-Type: application/json" \ -H "Authorization: Bearer sk-..." \ -d '{
    "prompt": "A cinematic shot of a futuristic city...",
    "webhook_url": "https://your-domain.com/webhook"
  }'

一个网关访问前沿模型

通过 ModelMax 统一调用文本、视觉、音频和视频模型。

47+

模型

12

供应商

4

能力

1

API 密钥

Gemini + Embedding

Gemini + Embedding

Google's most capable models, excelling at complex reasoning, deep multimodal understanding, and high-quality semantic embeddings.

能力
Gemini 3.1 Pro Preview
gemini-3.1-pro-preview
Chat2M tokens
Gemini 3 Pro Preview
gemini-3-pro-preview
Chat2M tokens
Gemini 3 Flash Preview
gemini-3-flash-preview
Chat1M tokens
Text Embedding 005
text-embedding-005
Embedding
Google Veo

Google Veo

State-of-the-art cinematic video generation with synchronized speech, sound effects, and prompt fidelity.

能力
Veo 3.1
veo-3.1
Video Gen
Veo 3.1 Fast
veo-3.1-fast
Video Gen
Veo 3
veo-3
Video Gen
Veo 3 Fast
veo-3-fast
Video Gen
OpenAI GPT

OpenAI GPT

OpenAI's latest GPT models for multimodal chat, coding, structured reasoning, and high-volume automation.

能力
GPT-5.5
gpt-5.5
Chat
GPT-5.4
gpt-5.4
Chat
GPT-5.4 Mini
gpt-5.4-mini
Chat
GPT-5.4 Nano
gpt-5.4-nano
Chat
Claude

Claude

Anthropic's Claude models with native Messages API support behind the unified OpenAI-compatible chat endpoint.

能力
Claude Opus 4.8
claude-opus-4-8
Chat
Claude Sonnet 4.6
claude-sonnet-4-6
Chat
Claude Haiku 4.5
claude-haiku-4-5
Chat
xAI Grok

xAI Grok

xAI's Grok models hosted through Google Cloud for advanced instruction following, synthesis, and fast high-volume text workflows.

能力
Grok 4.3
grok-4.3
Chat
Grok 4.1 Fast
grok-4.1-fast-non-reasoning
Chat
Kimi

Kimi

Moonshot's Kimi models combine Chinese-English reasoning, coding, and long-context capabilities for production workflows.

能力
Kimi K2.6
kimi-k2.6
Chat128K tokens
Kimi K2 Thinking
kimi-k2-thinking
Chat128K tokens
Kimi K2.5
kimi-k2.5
Chat128K tokens
MiniMax

MiniMax

Frontier large language model with strong reasoning, creativity, and highly consistent instruction-following.

能力
MiniMax M2
minimax-m2
Chat1M tokens
MiniMax M2.1
minimax-m2.1
Chat1M tokens
MiniMax M2.7
minimax-m2.7
Chat1M tokens
MiniMax M2.5
minimax-m2.5
Chat1M tokens
DeepSeek

DeepSeek

Top-tier open-source reasoning model with remarkable performance on STEM, coding, and mathematical benchmarks.

能力
DeepSeek R1
deepseek-r1
Chat128K tokens
DeepSeek V3.1
deepseek-v3.1
Chat128K tokens
DeepSeek V3.2
deepseek-v3.2
Chat128K tokens
Qwen

Qwen

Alibaba's robust MoE model excelling at code generation, logic, and comprehensive multilingual capabilities.

能力
Qwen3 Max
qwen3-max
Chat128K tokens
Qwen3.5 Plus
qwen3.5-plus
Chat128K tokens
Qwen3.5 Flash
qwen3.5-flash
Chat128K tokens
Qwen3 Coder 480B A35B
qwen3-coder-480b-a35b
Chat128K tokens
China DirectChina DirectChina DirectChina Direct

China Direct

Direct OpenAI-compatible access to GLM, Doubao, ERNIE, and Hunyuan models using provider API keys configured on the server.

能力
GLM-5.1
glm-5.1
Chat128K tokens
Doubao Seed 2.0 Pro
doubao-seed-2-0-pro-260215
Chat128K tokens
ERNIE 5.1
ernie-5.1
Chat128K tokens
Hunyuan 3 Preview
hy3-preview
Chat32K+ tokens

开发者优先的模型网关

ModelMax 将模型访问、成本控制和用量可观测性集中到一个控制台。

零加价

按供应商推理成本付费,清晰透明。

统一 API

用一个 OpenAI 兼容端点调用多家供应商模型。

用量分析

按日期、模型和成本跟踪请求。

开发者体验

兼容常见 SDK,迁移成本低。

Background pattern

生产级 AI 应用需要的统一入口

ModelMax 将模型访问、成本控制和用量可观测性集中到一个控制台。