Unlock exclusive introductory pricing for the newly launched Gemini 3.5 Flash.
Q

Qwen3.7 Plus

輸入:$0.32/M
輸出:$1.28/M
發布時間:Jun 8, 2026

Qwen3.7 Plus is a high-performance large language model developed by Alibaba Cloud. It supports long-context understanding up to 128K tokens, function calling, and multilingual tasks. Designed for complex reasoning, coding, and instruction-following scenarios.

新
商業用途

Qwen3.7 Plus 的 Playground

探索 Qwen3.7 Plus 的 Playground — 一個互動式環境，可測試模型並即時執行查詢。嘗試提示、調整參數，並立即迭代以加速開發並驗證使用案例。

Technical Specifications of Qwen3.7-Plus

ItemSpecification
Model NameQwen3.7-Plus
ProviderAlibaba Cloud (Qwen Team)
API Model IDqwen3.7-plus
Model TypeMultimodal agent model
Input TypesText, Images, Video
Output TypesText
Context WindowUp to 1,000,000 tokens
Maximum Input Tokens~991.8K
Maximum Output Tokens~65.5K
Core StrengthsVision-language reasoning, coding, GUI interaction, agent workflows
Built-in FeaturesFunction Calling, Structured Outputs, Cache, Web Search, Batch API
API CompatibilityOpenAI-compatible via DashScope / Model Studio

What Is Qwen3.7-Plus?

Qwen3.7-Plus is the balanced multimodal flagship in Alibaba's Qwen 3.7 generation. While Qwen3.7-Max focuses on pure text reasoning and long-horizon coding, Qwen3.7-Plus extends those capabilities with native image and video understanding, allowing it to reason across visual and textual information within a single model.

The model is designed around the idea of a multimodal interactive hybrid agent: it can interpret screenshots, analyze charts, understand interfaces, generate code from visual references, and use tools to complete multi-step tasks. This makes it particularly attractive for AI agents, GUI automation, and productivity workflows where visual context matters.

Main Features of Qwen3.7-Plus

  • Native multimodal input supporting text, images, and video in a unified reasoning pipeline.
  • Up to 1M-token context window, enabling large codebase analysis and long-document workflows.
  • Hybrid GUI + CLI agent capabilities, allowing the model to understand screens and interact with software environments.
  • Advanced coding and tool use, including function calling, structured outputs, and external tool integration.
  • Visual understanding for charts, documents, and screenshots, making it useful for business, engineering, and UI tasks.
  • OpenAI-compatible API endpoint, enabling relatively easy migration from existing LLM infrastructure.

Benchmark Performance of Qwen3.7-Plus

According to public benchmark reporting and third-party evaluation platforms, Qwen3.7-Plus delivers frontier-level multimodal and agentic performance:

  • Artificial Analysis Intelligence Index: 53.3.
  • GPQA Diamond: approximately 90.0%.
  • IFBench: approximately 78.0%.
  • AA Long Context Reasoning (AA-LCR): approximately 65.0%.
  • Terminal-Bench Hard: approximately 47.0%, reflecting strong agentic coding capabilities.

Alibaba also reports that Qwen3.7-Plus performs competitively with leading proprietary models on agent and coding benchmarks, with particular strength in multimodal tasks and UI interaction.

Qwen3.7-Plus vs Qwen3.7-Max vs Claude Opus 4.6

FeatureQwen3.7-PlusQwen3.7-MaxClaude Opus 4.6
Input ModalitiesText, Image, VideoText onlyText, Image
Context WindowUp to 1M tokens1M tokensLarge context
Vision UnderstandingExcellentNot supportedStrong
Agent / GUI InteractionNative focusLimitedGood
Long-Horizon Text ReasoningVery strongBest in Qwen familyExcellent
Best Use CaseMultimodal agents and productivityCoding, math, deep reasoningEnterprise reasoning and coding

For projects that involve screenshots, UI automation, visual debugging, or multimodal workflows, Qwen3.7-Plus is generally the better choice. For purely text-based reasoning or repository-scale coding, Qwen3.7-Max remains the stronger option.

Limitations

  • Qwen3.7-Plus is currently released as a preview-stage proprietary model, and some capabilities may evolve before stable release.
  • Function-calling and certain agent tooling features are still rolling out.
  • For purely text-only, reasoning-intensive workloads, Qwen3.7-Max may deliver slightly better performance.
  • As with most large multimodal models, developers should validate performance on their own image and UI datasets before production deployment.

Representative Use Cases

  1. AI agents that combine browser, GUI, and terminal interactions.
  2. Software debugging from screenshots and UI captures.
  3. Document, chart, and dashboard analysis.
  4. Visual coding assistants that generate code from mockups or diagrams.
  5. Enterprise productivity workflows involving mixed text and image inputs.
  6. Multimodal research assistants that combine web search with visual understanding

常見問題

Qwen3.7 Plus 的定價

探索 Qwen3.7 Plus 的競爭性定價，專為滿足各種預算和使用需求而設計。我們靈活的方案確保您只需為實際使用量付費，讓您能夠隨著需求增長輕鬆擴展。了解 Qwen3.7 Plus 如何在保持成本可控的同時提升您的專案效果。

ModelComet Price (USD / M Tokens)Official Price (USD / M Tokens)Discount
qwen3.7-plus
輸入:$0.32/M
輸出:$1.28/M
輸入:$0.4/M
輸出:$1.6/M
-20%

Qwen3.7 Plus 的範例程式碼和 API

存取完整的範例程式碼和 API 資源，以簡化您的 Qwen3.7 Plus 整合流程。我們詳盡的文件提供逐步指引，協助您在專案中充分發揮 Qwen3.7 Plus 的潛力。

#!/usr/bin/env bash

# Get your CometAPI key from https://www.cometapi.com/console/token
# Export it as: export COMETAPI_KEY="your-key-here"

response=$(curl --silent --location --request POST "https://api.cometapi.com/v1/responses" \
  --header "Authorization: Bearer $COMETAPI_KEY" \
  --header "Content-Type: application/json" \
  --header "Accept: application/json" \
  --data-raw '{
    "model": "qwen3.7-plus",
    "input": "你能做些什么？"
  }')

if command -v jq >/dev/null 2>&1; then
  printf '%s
' "$response" | jq -r '(
    [
      .output[]?
      | select(.type == "message")
      | .content[]?
      | select(.type == "output_text")
      | .text
    ][0]
  ) // .output_text // .'
else
  printf '%s
' "$response"
fi

cURL Code Example

#!/usr/bin/env bash

# Get your CometAPI key from https://www.cometapi.com/console/token
# Export it as: export COMETAPI_KEY="your-key-here"

response=$(curl --silent --location --request POST "https://api.cometapi.com/v1/responses" \
  --header "Authorization: Bearer $COMETAPI_KEY" \
  --header "Content-Type: application/json" \
  --header "Accept: application/json" \
  --data-raw '{
    "model": "qwen3.7-plus",
    "input": "你能做些什么？"
  }')

if command -v jq >/dev/null 2>&1; then
  printf '%s\n' "$response" | jq -r '(
    [
      .output[]?
      | select(.type == "message")
      | .content[]?
      | select(.type == "output_text")
      | .text
    ][0]
  ) // .output_text // .'
else
  printf '%s\n' "$response"
fi

Python Code Example

from openai import OpenAI
import os

# Get your CometAPI key from https://www.cometapi.com/console/token
client = OpenAI(
    api_key=os.environ["COMETAPI_KEY"],
    base_url="https://api.cometapi.com/v1",
)

response = client.responses.create(
    model="qwen3.7-plus",
    input="你能做些什么？",
)

print(response.output_text)

JavaScript Code Example

import OpenAI from "openai";

// Get your CometAPI key from https://www.cometapi.com/console/token
const client = new OpenAI({
  apiKey: process.env.COMETAPI_KEY,
  baseURL: "https://api.cometapi.com/v1",
});

const response = await client.responses.create({
  model: "qwen3.7-plus",
  input: "你能做些什么？",
});

console.log(response.output_text);

Qwen3.7 Plus的版本

Qwen3.7 Plus擁有多個快照的原因可能包括：更新後輸出結果存在差異需保留舊版快照以確保一致性、為開發者提供適應與遷移的過渡期，以及不同快照對應全球或區域端點以優化使用者體驗等潛在因素。各版本間的具體差異請參閱官方文件說明。

Version
qwen3.7-plus