模型定价企业
500+ AI 模型 API,一次搞定,就在 CometAPI
模型 API
开发者
快速入门文档API 仪表板
公司
关于我们企业
资源
AI 模型博客更新日志支持
服务条款隐私政策
© 2026 CometAPI · All rights reserved
Home/Models/Flux/flux-dev
F

flux-dev

每次请求:$0.03
商用
概览
功能亮点
定价
API

Technical Specifications of flux-dev

SpecificationDetails
Model IDflux-dev
Model familyFLUX.1 [dev]
Provider / originBlack Forest Labs
ModalityText-to-image generation
Architecture12B-parameter rectified flow transformer
Primary use caseGenerating high-quality images from natural-language prompts
Performance positioningDesigned as a high-quality open-weight variant positioned below FLUX.1 [pro] and above faster lightweight variants in the FLUX family
Prompt handlingStrong prompt adherence and competitive instruction following for complex image descriptions
Efficiency approachTrained with guidance distillation for more efficient inference
Availability patternCommonly available through hosted inference APIs and model platforms, with open weights published for the base model ecosystem
Licensing noteThe original FLUX.1 [dev] release is associated with a non-commercial license from Black Forest Labs, so downstream usage terms should be checked carefully in the environment where you deploy it

What is flux-dev?

flux-dev is CometAPI’s model identifier for FLUX.1 [dev], a text-to-image model from Black Forest Labs built to generate detailed, prompt-faithful images from written descriptions. Black Forest Labs describes FLUX.1 [dev] as a 12 billion parameter rectified flow transformer, and positions it as a high-quality model that is second to FLUX.1 [pro] within the original FLUX.1 lineup.

In practice, flux-dev is used when you want stronger image quality and better prompt adherence than lightweight image generators, while still benefiting from the broader FLUX ecosystem. The model is widely recognized for producing polished outputs, handling stylistic directions well, and supporting research and creative workflows through open model availability on major platforms.

Because FLUX.1 [dev] sits in a family that later expanded with tool-oriented and editing-oriented variants, it is best understood as a core text-to-image foundation model: you give it a text prompt, and it returns one or more generated images matching the requested scene, style, composition, or concept.

Main features of flux-dev

  • High-quality image generation: FLUX.1 [dev] is positioned by Black Forest Labs as a cutting-edge text-to-image model with output quality just below FLUX.1 [pro], making it suitable for professional-looking concept art, illustrations, marketing visuals, and creative prototyping.
  • Strong prompt adherence: The model is specifically noted for competitive prompt following, which helps when prompts include multiple objects, layout cues, stylistic instructions, or detailed scene descriptions.
  • 12B rectified flow transformer architecture: Its 12-billion-parameter rectified flow transformer design is a core part of the model’s capability profile and is one reason it is associated with strong fidelity and image detail.
  • Guidance-distilled efficiency: Black Forest Labs states that FLUX.1 [dev] was trained using guidance distillation, which improves efficiency compared with less optimized generation approaches.
  • Open ecosystem compatibility: The model has been released in forms that support broader experimentation, deployment, and workflow customization, including ecosystem support through Hugging Face, ONNX exports, and related FLUX tooling.
  • Flexible creative workflow support: flux-dev is well suited for ideation, visual drafting, style exploration, and pipeline integration where developers want a strong general-purpose text-to-image model rather than a narrowly specialized generator. This is an inference based on the official positioning and available deployment formats.

How to access and integrate flux-dev

Step 1: Sign Up for API Key

To access flux-dev, first register on CometAPI and generate your API key from the dashboard. After signing in, create a key for your application and store it securely in an environment variable or secret manager. You will use this key to authenticate every request to the flux-dev API.

Step 2: Send Requests to flux-dev API

Once you have your API key, send requests to the CometAPI endpoint with flux-dev as the model name.

curl --request POST \
  --url https://api.cometapi.com/v1/images/generations \
  --header "Authorization: Bearer $COMETAPI_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "model": "flux-dev",
    "prompt": "A cinematic sci-fi city at sunset, ultra-detailed, volumetric lighting"
  }'

Python example:

import os
import requests

url = "https://api.cometapi.com/v1/images/generations"
headers = {
    "Authorization": f"Bearer {os.environ['COMETAPI_API_KEY']}",
    "Content-Type": "application/json",
}
payload = {
    "model": "flux-dev",
    "prompt": "A cinematic sci-fi city at sunset, ultra-detailed, volumetric lighting",
}

response = requests.post(url, headers=headers, json=payload, timeout=120)
response.raise_for_status()
print(response.json())

Step 3: Retrieve and Verify Results

After submission, CometAPI returns the generated result payload for flux-dev. Parse the response to retrieve image URLs, file data, or related metadata depending on your integration pattern. You should then verify that the output matches the prompt, check for application-level safety or policy compliance, and log request IDs or response metadata for debugging, auditing, and workflow reliability.

flux-dev 的功能

了解 flux-dev 的核心能力,帮助提升性能与可用性,并改善整体体验。

flux-dev 的定价

查看 flux-dev 的竞争性定价,满足不同预算与使用需求,灵活方案确保随需求扩展。
Comet 价格 (USD / M Tokens)官方定价 (USD / M Tokens)折扣
每次请求:$0.03
每次请求:$0.0375
-20%

flux-dev 的示例代码与 API

获取完整示例代码与 API 资源,简化 flux-dev 的集成流程,我们提供逐步指导,助你发挥模型潜能。

更多模型

G

Nano Banana 2

输入:$0.4/M
输出:$2.4/M
核心能力概览:分辨率:最高 4K(4096×4096),与 Pro 相当。参考图像一致性:最多 14 张参考图像(10 个物体 + 4 个角色),保持风格/角色一致性。极端纵横比:新增 1:4、4:1、1:8、8:1 比例,适用于长图、海报和横幅。文本渲染:高级文本生成,适用于信息图和营销海报版式。搜索增强:集成 Google Search + 图片搜索。Grounding:内置思维过程;生成前先对复杂提示进行推理。
C

Claude Opus 4.7

输入:$4/M
输出:$20/M
用于智能体和编程的最智能模型
C

Claude Opus 4.6

输入:$4/M
输出:$20/M
Claude Opus 4.6 是 Anthropic 的“Opus”级大型语言模型,于 2026 年 2 月发布。其定位为知识工作与研究工作流的主力模型——提升长上下文推理、多步骤规划、工具使用(包括代理型软件工作流),以及计算机使用类任务,如自动生成幻灯片和电子表格。
A

Claude Sonnet 4.6

输入:$2.4/M
输出:$12/M
Claude Sonnet 4.6 是迄今为止我们最强大的 Sonnet 模型。它对模型在编码、计算机使用、长上下文推理、智能体规划、知识工作和设计等方面的能力进行了全面升级。Sonnet 4.6 还在 beta 阶段提供 1M token 上下文窗口。
O

GPT-5.4 nano

输入:$0.16/M
输出:$1/M
GPT-5.4 nano 专为速度和成本最为关键的任务而设计,例如分类、数据提取、排序以及子智能体。
O

GPT-5.4 mini

输入:$0.6/M
输出:$3.6/M
GPT-5.4 mini 将 GPT-5.4 的优势融入到一款更快速、更高效、专为大规模工作负载设计的模型中。