Ultimate AI API Gateway

Call all mainstream large models easily via a fast, stable, unified interface. No time limits, pay-as-you-go, transparent billing with no hidden charges. After an online top-up, you can use all models; already serving tens of thousands of users.

Documentation Center Try now

Full coverage of mainstream models

Unified API, easy model switching

from openai import OpenAI

client = OpenAI(

base_url="https://tokensmind.ai",

api_key="sk-***"

)

response = client.chat.completions.create(

model="gpt-4o",

Gemini 3 Pro Preview / DeepSeek V3 / Claude Opus 4.6 ...

)

Unlimited concurrency, always online

OpenAI

Anthropic

Google

ByteDance

Qwen

Kimi

MiniMax

DeepSeek

Zhipu AI

Common App Setup Guides

Cursor

Cursor, the AI coding editor, connects to TokensMind to deliver semantic completion, debugging, and collaboration.

Open guide

Claude Code

Anthropic’s terminal coding agent; with TokensMind, call Claude models for edits, reviews, and refactors.

Open guide

Open Claw

For agent and automation workflows; route model calls through TokensMind’s unified API.

Open guide

Qwen Code

Alibaba’s Qwen coding assistant; use Qwen models in your terminal or IDE via TokensMind.

Open guide

OpenAI Codex

OpenAI’s coding assistant and CLI; access GPT-family models through TokensMind with one API.

Open guide

热门模型

精选各类目顶级模型，几分钟即可开始接入。

LLM图片Video

claude-opus-4-8

No model description

Pay as you go

gpt-5.5

GPT-5.5 is a frontier upgrade in the GPT-5 family for complex professional work, further improving reasoning, coding, instruction following, and reliability.

Pay as you go

deepseek-v4-pro

DeepSeek-V4 Pro is the high-performance DeepSeek V4 model, strengthening complex reasoning, coding, analysis, and agentic workflows for production use.

Pay as you go

deepseek-v4-flash

DeepSeek-V4 Flash is the fast version of the DeepSeek V4 series, designed for low-latency, high-throughput chat, coding assistance, and lightweight reasoning.

Pay as you go

kimi-k2.6

Kimi K2.6 is an upgraded Kimi K2 model with stronger long-context, coding, agentic, and multimodal understanding capabilities for complex production tasks.

Pay as you go

qwen3.5-plus

Qwen3.5 native vision-language Plus model built on a hybrid architecture that combines linear attention with sparse MoE for higher inference efficiency. Across many benchmarks, the 3.5 series reaches performance comparable to top frontier models, with major gains over the 3 series in both text-only and multimodal tasks. This release is functionally equivalent to the qwen3.5-plus-2026-02-15 snapshot.

TextImageVideo

glm-5.1

GLM-5.1 is Zhipu's latest flagship model, with major improvements in coding and long-horizon tasks, able to plan, execute, and iterate autonomously for extended work.

Pay as you go

MiniMax-M2.7

MiniMax M2.7 is MiniMax's next-generation general model with stronger coding, tool use, office productivity, and complex task handling.

Pay as you go

gpt-image-2

GPT-Image-2 is OpenAI's next-generation image generation and editing model, improving image quality, instruction following, text rendering, and iterative editing.

Pay as you go

gemini-3.1-flash-image-preview

Gemini 3.1 Flash Image Preview (officially marketed as Nano Banana 2) offers professional-grade visual intelligence with lightning-fast efficiency and grounded generation. It is the speed-optimized counterpart to Gemini 3 Pro Image, designed for scale and broad developer use cases.

TextImageReasoning

gemini-3-pro-image-preview

Gemini-3-Pro-Image-Preview (Nano Banana Pro) is a high-performance image generation and editing model built on Gemini 3 Pro. It offers stronger multimodal understanding and grounded semantic reasoning, generates structured visual content quickly, supports search-based real-time knowledge grounding, and performs especially well on text rendering, multi-image consistency, and identity preservation.

TextImageReasoning

gemini-2.5-flash-image

Gemini 2.5 Flash Image (Nano-Banana) is an advanced image generation and editing model that can naturally merge multiple images into a single composition while keeping characters consistent across iterative creation. It supports precise local image edits through natural language and combines strong world knowledge with image generation and editing, making it a strong fit for creative design, content production, advertising, and visual storytelling.

TextImage

wan2.6-i2v-flash

Wan 2.6 Image-to-Video Flash delivers faster generation at a better price-performance ratio. Its intelligent shot scheduling supports multi-shot storytelling, stable multi-character dialogue, more natural voice timbre, and video generation up to 15 seconds.

Pay per view

viduq3-pro

Vidu Q3 Pro is Vidu's high-quality video generation model with stronger visual quality, motion consistency, and complex shot control.

Pay as you go

wan2.6-r2v-flash

Wan 2.6 Reference-to-Video Flash generates faster at a better price-performance ratio. It supports referencing specific people or arbitrary objects to preserve appearance and voice consistency, including multi-character co-performance scenarios.

Pay per view

viduq3-turbo

Vidu Q3 Turbo is the fast Vidu Q3 variant, focused on faster generation and better cost efficiency for video creation.

Pay as you go

MiniMax-Hailuo-2.3

MiniMax Hailuo 2.3 is MiniMax's video generation model for high-quality text/image-to-video creation, suited to cinematic storytelling, ads, and creative content.

Pay as you go

查看全部模型

3-Minute Quick Integration

Start using hundreds of AI models in minutes

Sign up and create a key

Create an account and generate your API key in the dashboard. No credit card required.

Get API Key View Docs

Update configuration

Change the API endpoint to TokensMind. OpenAI SDK users only need to update one line of code.

View Docs

Start calling models

Access hundreds of AI models, including GPT-5, Claude, and Sora, through one unified API.

curl https://api.tokensmind.com/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <YOUR_API_KEY>" \
  -d '{
    "model": "gpt-5",
    "messages": [
      {
        "role": "user",
        "content": "Describe what TokensMind can do in one sentence."
      }
    ]
  }'

Limited-Time EventWorld Cup AI Prediction

AI World Cup Oracle

Predict match results, generate your own AI poster, and share your World Cup take.

Join Now