模型目錄
瀏覽 ReachAPI 目前 41 個官方正版 AI API，涵蓋 OpenAI、Google、Anthropic 等廠商的文本、圖像與影片生成模型，支援比較價格並快速開始開發。
共 39 個結果
圖像Bytedance
可用Seedream 5.0 Pro is designed for image workflows that need precise control over subjects and placement. It supports text-to-image, single-image editing, up to 10 reference images, and interactive edits driven by freeform marks, points, or bounding boxes.
圖像Bytedance
可用Seedream 5.0 Lite is a cost-efficient image generation and editing model designed for enterprise visual production, combining strong prompt reasoning, multimodal inputs, and online retrieval support for grounded creative work.
圖像Bytedance
可用Seedream 4.5 is designed around a unified generation and editing architecture, supporting high-quality image creation, multi-image editing consistency, subject detail preservation, and polished outputs up to roughly 4MP.
圖像OpenAI
可用GPT Image 2 Async exposes GPT Image 2 through ReachAPI's async image task interface, designed for queued generation, task polling, terminal callbacks, and final image URLs returned in `data[].url`.
圖像OpenAI
可用GPT Image 2 is OpenAI's latest image model, designed for fast, high-quality image generation and editing with flexible sizes, strong instruction following, and high-fidelity image input support.
圖像Nano Banana maps to Gemini 2.5 Flash Image and is built for speed and efficiency, making it a strong fit for high-volume, low-latency image tasks where throughput matters most.
MiniMax
可用MiniMax Hailuo-03 — high-quality text/image-to-video generation (4-15s, 2K or 768P resolution).
Bytedance
可用Seedance 2 Fast provides the Seedance 2 creative workflow for faster, lower-cost video tasks with 4-15 second outputs, 480p/720p tiers, and token pricing by video input.
Bytedance
可用Seedance 2 is the full Seedance 2 video model on ReachAPI, supporting text, image, video, and audio reference workflows with 4-15 second outputs and token pricing by resolution and video input.
Bytedance
可用Seedance 1.5 Pro is an asynchronous video generation model on ReachAPI for 4-12 second outputs, adaptive aspect ratios, and token pricing that varies by audio output.
Bytedance
可用Seedance 1.0 Pro Fast is the faster, lower-cost Seedance 1.0 option for 2-12 second video generation with official-style token-based pricing.
文本Moonshot
可用Kimi K3 is Kimi's most capable flagship model, combining always-on reasoning, native visual understanding, a 1M-token context window, and advanced tool use for long-horizon coding and knowledge work.
文本Anthropic
可用Claude Opus 5 is Anthropic's flagship model for complex agentic coding and enterprise work, with stronger deep reasoning, long-horizon execution, tool use, and multi-agent coordination.
文本OpenAI
可用GPT-5.6 Luna is the efficient tier in the GPT-5.6 family. It provides reasoning, tool use, and a 1.05M-token context window for teams that need scalable throughput and lower unit costs.
文本OpenAI
可用GPT-5.6 Terra is the balanced tier in the GPT-5.6 family. It offers strong reasoning, tool use, and a 1.05M-token context window for enterprise workloads that need a practical balance of capability and cost.
文本OpenAI
可用GPT-5.6 Sol is the frontier tier in the GPT-5.6 family. It pairs the family's strongest capability with a 1.05M-token context window for complex enterprise workflows that need reliable reasoning, tool use, and long-context execution.
文本Anthropic
可用Claude Sonnet 5 is Anthropic's newest Sonnet model, built for long-context reasoning, agentic coding, tool-heavy workflows, and production workloads that need stronger sustained performance than prior Sonnet releases.
