/ models

126 models,one endpoint.

Curated model records are wired into this public marketplace with price, context, capability, routing status, and source labels. Shortlist the workload first, then run candidates through one OpenAI-compatible endpoint.

routing candidates123/126
Claude Fable 5.1$10/1M
Claude Haiku 4.5$1/1M
Claude Opus 4.5$5/1M
Claude Opus 4.6$5/1M
16providers1sources2Mmax context$0lowest input
Reset

126 of 126 models

Model cards with source labels and copyable OpenAI-compatible calls.

What can you compare on this page?

The NextModel marketplace compares model provider, input price, output price, context length, latency estimate, capabilities, use cases, availability, routing status, and source label so teams can shortlist candidates before sending production traffic.

AnthropicCatalog

面向长周期 Agent、大型代码工程、多步科研推理、重度知识工作的前沿大模型,在 Fable5 基础上重点强化长时间自主任务稳定性、工具调用、科学数学推理、代码全库开发能力,支持多小时持续 Agent 会话,具备自我校验纠错能力。

$10 / 1M tokensInput$50 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

2.4万亿参数MoE旗舰,编程与办公能力全面跃升,可自主编程十数天交付完整项目。胜任法律、金融、设计等数百种专业任务,一次对话端到端交付生产级成果。原生视觉理解贯穿规划、执行与验证全流程,支持超长文档与长视频的深度语义解析。长程任务中自主规划与闭环迭代,持续进化。

$2 / 1M tokensInput$6 / 1M tokensOutput1MContext
Best forGeneral chat via Qwen, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
Black Forest LabsCatalog

Available through the NextModel gateway via Black Forest Labs.

$0.2 / imagePer imageContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
Black Forest LabsCatalog

Available through the NextModel gateway via Black Forest Labs.

$0.075 / imagePer imageContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

OpenAI GPT Image 系列标准模型,支持文本和图像输入。擅长添加、删除、组合和混合元素编辑,保持构图和光照一致性,适合创意设计和图像处理。 官方API参考:

$0.26 / imagePer imageContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

OpenAI GPT Image 系列轻量模型,性价比极高。保持良好的图像生成质量,适合大批量图像生成和对成本敏感的场景。 官方API参考:

$0.0539 / imagePer imageContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

OpenAI 最新图像生成模型,支持文本和图像输入。生成速度比 GPT Image 1 快 4 倍,成本降低 20%。精确编辑控制,出色的文字渲染能力,支持密集小字和复杂排版,适合专业设计和营销素材制作。 官方API参考:

$0.218 / imagePer imageContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

OpenAI 新一代推理式多模态图像生成模型,支持文生图/图生图/对话式编辑,具备思考模式、精准多语言文字渲染、2K分辨率、批量连贯出图能力,专为商用级视觉内容生产优化。 官方API参考:

$0.221 / imagePer imageContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

Available through the NextModel gateway via OpenAI.

$0.02 / 1M tokensInput$0 / 1M tokensOutputContext
Best forGeneral chat via OpenAI, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

Available through the NextModel gateway via OpenAI.

$0.13 / 1M tokensInput$0 / 1M tokensOutputContext
Best forGeneral chat via OpenAI, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
Z AiCatalog

GLM-5.3 是智谱最新旗舰模型,1M长上下文,支持思考长度控制,在复杂软件工程与 Agent 任务能力全面进阶。它使用与 GLM-5.2 相同的基础模型——所有提升均来自后训练。与 GLM-5.2 相比,它在复杂编程和长程任务方面表现更加出色。

$1.40 / 1M tokensInput$4.40 / 1M tokensOutput1MContext
Best forGeneral chat via Z Ai, API workloads
RoutingConfigured
StreamingTool callingJSON modeLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-5 Codex 优化版,专为代码生成和软件工程任务优化。支持 Responses API,擅长修复 Bug、代码编辑和复杂代码库问答。

$1.25 / 1M tokensInput$10 / 1M tokensOutput400kContext
Best forGeneral chat via OpenAI, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Qwen3系列小尺寸视觉理解模型,实现思考模式和非思考模式的有效融合,效果优于开源版Qwen3-VL-30B-A3B,响应速度快。全面升级图像/视频理解,支持长视频长文档等超长上下文、空间感知与万物识别;具备视觉2D/3D定位能力,胜任复杂现实任务。

Starting at $0.03 / 1M tokensInput$0.22 / 1M tokensOutput256kContext
Best forGeneral chat via Qwen, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Qwen3.5原生视觉语言系列Flash模型,基于混合架构设计,融合了线性注意力机制与稀疏混合专家模型,实现了更高的推理效率。模型效果在纯文本与多模态方面相较3系列均实现飞跃式进步;响应速度快,兼具推理速度和性能。

Starting at $0.03 / 1M tokensInput$0.29 / 1M tokensOutput1MContext
Best forGeneral chat via Qwen, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Qwen3系列小尺寸视觉理解模型,实现思考模式和非思考模式的有效融合,效果优于开源版Qwen3-VL-30B-A3B,响应速度快。全面升级图像/视频理解,支持长视频长文档等超长上下文、空间感知与万物识别;具备视觉2D/3D定位能力,胜任复杂现实任务。

Starting at $0.05 / 1M tokensInput$0.4 / 1M tokensOutput256kContext
Best forGeneral chat via Qwen, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Qwen3.5原生视觉语言系列Plus模型,基于混合架构设计,融合了线性注意力机制与稀疏混合专家模型,实现了更高的推理效率。在多项任务评测中,3.5系列均展现出与当前顶尖前沿模型相媲美的卓越性能,模型效果在纯文本与多模态方面相较3系列均实现飞跃式进步。

Starting at $0.12 / 1M tokensInput$0.69 / 1M tokensOutput1MContext
Best forGeneral chat via Qwen, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Qwen3系列视觉理解模型,实现思考模式和非思考模式的有效融合,视觉智能体能力在OS World等公开测试集上达到世界顶尖水平。此版本在视觉coding、空间感知、多模态思考等方向全面升级;视觉感知与识别能力大幅提升,支持超长视频理解。

Starting at $0.15 / 1M tokensInput$1.44 / 1M tokensOutput256kContext
Best forGeneral chat via Qwen, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Qwen3.6原生视觉语言系列Flash模型,模型效果相较3.5-Flash显著提升。本模型重点提升agentic coding能力(在多项代码智能体基准上大幅超越前代)、数学推理和代码推理能力;视觉方面在空间智能能力上显著增强,物体定位与目标检测提升尤为突出。

Starting at $0.25 / 1M tokensInput$1.50 / 1M tokensOutput1MContext
Best forGeneral chat via Qwen, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

千问3系列Max模型,相较preview版本在智能体编程与工具调用方向进行了专项升级。本次发布的正式版模型达到领域SOTA水平,适配场景更加复杂的智能体需求。

Starting at $0.36 / 1M tokensInput$1.44 / 1M tokensOutput256kContext
Best forGeneral chat via Qwen, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Qwen3系列视觉理解模型,实现思考模式和非思考模式的有效融合,视觉智能体能力在OS World等公开测试集上达到世界顶尖水平。此版本在视觉coding、空间感知、多模态思考等方向全面升级;视觉感知与识别能力大幅提升,支持超长视频理解。

Starting at $0.2 / 1M tokensInput$1.60 / 1M tokensOutput256kContext
Best forGeneral chat via Qwen, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Qwen3.6原生视觉语言系列Plus模型,展现出与当前顶尖前沿模型相媲美的卓越性能,模型效果相较3.5系列显著提升。模型在Agentic coding、前端编程、Vibe coding等代码能力、多模态万物识别、OCR、物体定位等能力上显著增强。

Starting at $0.5 / 1M tokensInput$3 / 1M tokensOutput1MContext
Best forGeneral chat via Qwen, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

千问3系列Max模型,相较preview版本在智能体编程与工具调用方向进行了专项升级。本次发布的正式版模型达到领域SOTA水平,适配场景更加复杂的智能体需求。

Starting at $1.20 / 1M tokensInput$6 / 1M tokensOutput256kContext
Best forGeneral chat via Qwen, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Available through the NextModel gateway via Qwen.

Starting at $1.30 / 1M tokensInput$7.80 / 1M tokensOutput256kContext
Best forGeneral chat via Qwen, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
DeepSeekCatalog

DeepSeek-V4-Pro-0813核心强化Agent智能体能力,在代码智能体、工具调用、长链路任务执行等公开基准测试中性能提升,依托DeepSeek Harness极简测试框架完成专项优化,更适配自动化开发、复杂任务编排场景。

$1.33 / 1M tokensInput$3.98 / 1M tokensOutput1MContext
Best forChinese Q&A, general chat
RoutingConfigured
StreamingTool callingJSON modeLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
VolcengineCatalog

高效轻量型模型,专为低时延、高并发与成本敏感场景设计,追求极致性价比,适合边缘部署与大规模基础对话应用。

Starting at $0.03 / 1M tokensInput$0.3 / 1M tokensOutput256kContext
Best forChinese Q&A, general chat
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
VolcengineCatalog

字节跳动旗舰级通用 Agent 模型,侧重长链路推理与复杂任务稳定性,多模态理解全面升级,适配科研分析、专业方案评审等高端复杂场景。

Starting at $0.48 / 1M tokensInput$2.36 / 1M tokensOutput256kContext
Best forChinese Q&A, general chat
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
VolcengineCatalog

均衡型通用模型,兼顾性能与成本,综合能力超越上一代主力模型,适配 95% 企业高频场景,是多数业务的性价比之选。

Starting at $0.09 / 1M tokensInput$0.53 / 1M tokensOutput256kContext
Best forChinese Q&A, general chat
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
VolcengineCatalog

编程专用模型,深度优化代码生成、审查与调试能力,与 TRAE 工具链无缝协同,大幅提升开发者编码效率与质量。

Starting at $0.48 / 1M tokensInput$2.36 / 1M tokensOutput256kContext
Best forChinese Q&A, general chat
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Qwen3.5原生视觉语言系列Flash模型,基于混合架构设计,融合了线性注意力机制与稀疏混合专家模型,实现了更高的推理效率。模型效果在纯文本与多模态方面相较3系列均实现飞跃式进步;响应速度快,兼具推理速度和性能。

$0.1 / 1M tokensInput$0.4 / 1M tokensOutput1MContext
Best forGeneral chat via Qwen, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
X AiCatalog

Grok 4.1 快速非推理是 xAI 前沿工具调用模型的即时响应变体。它跳过思考令牌阶段,以实现最高速度和低延迟,同时提供针对代理式工作流优化的高质量输出。它基于 Grok 4.1 的传承,保持了强大的工具调用准确性,相比前几代减少了幻觉,并支持较大的上下文窗口。该变体在高通量、实时场景中表现出色,优先考虑即时响应而非逐步推理。

$0.2 / 1M tokensInput$0.5 / 1M tokensOutput2MContext
Best forGeneral chat via X Ai, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
X AiCatalog

Grok 4.2 是 xAI 最新的大型语言模型,专为强推理、多模态理解和企业应用而构建。它提升了指令跟踪、诚信性和校准能力,相较早期版本,同时支持单代理和多代理工作流程。Grok 4.2 被设计为通用的真理追寻助手,在配合适当防护下,非常适合研究、分析、编码及复杂专业任务。

$1.25 / 1M tokensInput$2.50 / 1M tokensOutput256kContext
Best forGeneral chat via X Ai, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
X AiCatalog

Grok 4.2 是 xAI 最新的大型语言模型,专为强推理、多模态理解和企业应用而构建。它提升了指令跟踪、诚信性和校准能力,相较早期版本,同时支持单代理和多代理工作流程。Grok 4.2 被设计为通用的真理追寻助手,在配合适当防护下,非常适合研究、分析、编码及复杂专业任务。

$1.25 / 1M tokensInput$2.50 / 1M tokensOutput256kContext
Best forGeneral chat via X Ai, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
VolcengineCatalog

迈向生产级智能的新一代大模型,全面升级 Coding、Agent 与多模态能力,以更强的自主规划、长链路执行和动态修复能力,胜任企业真实复杂任务。

$0.89 / 1M tokensInput$4.42 / 1M tokensOutput256kContext
Best forChinese Q&A, general chat
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Available through the NextModel gateway via Qwen.

$2.50 / 1M tokensInput$7.50 / 1M tokensOutput1MContext
Best forGeneral chat via Qwen, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
VolcengineCatalog

效果与成本均衡,全面升级 Coding、Agent 与多模态能力,以更强的自主规划、长链路执行和动态修复能力,胜任企业真实复杂任务。

$0.45 / 1M tokensInput$2.21 / 1M tokensOutput256kContext
Best forChinese Q&A, general chat
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
X AiCatalog

Grok 4.1 快速推理是一个专注于高性能代理工具调用的前沿多模态模型。它能够准确快速地推理并完成代理任务,在客户支持和财务等复杂现实场景中表现出色。配合代理工具,它使开发者能够构建专注于工具调用和代理搜索的生产级代理。它拥有更自然流畅的对话,同时保持了强大的核心推理能力,更敏锐地捕捉细腻的意图,引人入胜,个性连贯。

$0.2 / 1M tokensInput$0.5 / 1M tokensOutput2MContext
Best forGeneral chat via X Ai, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

2.4万亿参数MoE旗舰,编程与办公能力全面跃升,可自主编程十数天交付完整项目。胜任法律、金融、设计等数百种专业任务,一次对话端到端交付生产级成果。原生视觉理解贯穿规划、执行与验证全流程,支持超长文档与长视频的深度语义解析。长程任务中自主规划与闭环迭代,持续进化。

$1.77 / 1M tokensInput$5.30 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
DeepSeekCatalog

高效轻量化MoE模型,总参284B,激活13B,原生支持百万超长上下文能力。推理速度快、延迟低、调用成本低廉,综合能力均衡,主打高并发、轻量化任务,适合日常对话、内容创作、基础 RAG、批量文案处理等普惠刚需场景。

$0.45 / 1M tokensInput$1.33 / 1M tokensOutput1MContext
Best forChinese Q&A, general chat
RoutingConfigured
StreamingTool callingJSON modeLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Qwen3.7原生视觉语言系列Flash模型,相较3.6-Flash全面提升多模态理解与Agent执行能力。重点强化多模态基础能力、万物识别能力更强,真实世界感知与空间智能进一步提升,Search Agent、CI Agent等多模态Agent场景能力显著升级、端到端任务执行更稳定,多模态Coding能力优化、vibe coding 体验更加流畅。

Starting at $0.03 / 1M tokensInput$0.12 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
GoogleCatalog

Gemini 3.5 Flash-Lite 是 Gemini 3 系列高性能原生多模态推理模型的新增成员。该模型成本效益突出、响应速度快,专为翻译、分类这类高吞吐量、时延敏感型任务优化,同时支持智能体工作流。

$0.3 / 1M tokensInput$2.50 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
GoogleCatalog

Gemini 3.8 Flash 是现阶段综合性能最为出色的主力模型。相较 Gemini 3.7 Flash,该模型在软件工程、智能代理任务以及专业领域复杂多步骤核心推理能力方面,均实现全方位大幅升级。

$0.75 / 1M tokensInput$3.75 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-5.6 系列极速高性价比轻量模型,全系列最低调用成本、低延迟高吞吐,面向实时客服、批量摘要、内容清洗、高频并发标准化企业任务场景

Starting at $0.2 / 1M tokensInput$1.20 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-5.6 均衡中端通用模型,兼顾强推理能力与减半使用成本,适配企业日常文档处理、业务分析、常规开发与批量知识运维,性能对标上代旗舰 GPT-5.5

Starting at $2 / 1M tokensInput$12 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-5.6 Sol is the flagship model in OpenAI's GPT-5.6 series. It is suited for complex reasoning, coding, and agentic workflows, and is particularly strong at command-line and multi-step coding tasks...

Starting at $5 / 1M tokensInput$30 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Qwen3.5原生视觉语言系列Plus模型,基于混合架构设计,融合了线性注意力机制与稀疏混合专家模型,实现了更高的推理效率。在多项任务评测中,3.5系列均展现出与当前顶尖前沿模型相媲美的卓越性能,模型效果在纯文本与多模态方面相较3系列均实现飞跃式进步。

Starting at $0.4 / 1M tokensInput$2.40 / 1M tokensOutput1MContext
Best forGeneral chat via Qwen, API workloads
RoutingConfigured
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
MoonshotaiCatalog

kimi-k2.7-code 是 Kimi 迄今为止最智能的编码模型。它能更可靠地在长时间上下文中执行指令,并以更高的成功率完成编程任务。它支持文本、图片和视频输入,以及思考模式、对话和客服任务。

$0.9 / 1M tokensInput$3.72 / 1M tokensOutput256kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
Z AiCatalog

GLM-5.2是智谱的旗舰模型,适用于长时程、项目规模的任务。它拥有真正可用的1M-token上下文窗口,能够处理大型工程上下文,在扩展工作流程中保持一致性,并以更高的可靠性执行长时间运行的任务。它专为端到端软件开发而设计,从需求分析、架构规划到实现、测试、调试和多平台部署,非常适合自主代理和复杂的工程工作流程。

$1.18 / 1M tokensInput$4.12 / 1M tokensOutput1MContext
Best forGeneral chat via Z Ai, API workloads
RoutingConfigured
StreamingTool callingJSON modeLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Qwen3.7系列中高性价比Plus模型,在强大文本能力的基础上全面升级了视觉-语言能力,同时保持了在编码、工具使用和生产力工作流方面的完整智能体能力。其核心特色为多模态交互混合智能体能力,能够感知真实世界场景、读取屏幕并操作 GUI、基于视觉参考生成代码、端到端导航移动应用。

Starting at $0.3 / 1M tokensInput$1.18 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Qwen3.7-Max is the flagship model in Alibaba's Qwen3.7 series. It supports text input and output and is designed for agent-centric workloads, with particular strengths in coding, office and productivity tasks,...

$1.65 / 1M tokensInput$4.96 / 1M tokensOutput1MContext
Best forGeneral chat via Qwen, API workloads
RoutingConfigured
StreamingTool callingJSON modeLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
GoogleCatalog

Google Gemini 3.1 Flash-Lite,超轻量快速模型。支持文本、图像、视频多模态输入,极低成本。适合对延迟和成本极度敏感的大规模部署场景,支持缓存进一步降本。

$0.25 / 1M tokensInput$1.50 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-Chat Latest 是 OpenAI 自动追踪更新的通用聊天模型别名,对应当前最新 GPT Instant 版本,拥有 400K 超大上下文、多模态图文输入能力,幻觉更低、工具调用更稳定,适配通用对话、检索增强与轻量智能体开发场景

$5 / 1M tokensInput$30 / 1M tokensOutput400kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
X AiCatalog

xAI 新一代智能体旗舰大模型,标配百万级上下文与原生多模态,内置持续推理与工具调用能力,推理、代码、长文档处理综合性能大幅升级,API 调用成本大幅下调,兼顾复杂任务处理与高性价比,适配自动化工作流、图文解析、实时资讯分析等场景。

$1.25 / 1M tokensInput$2.50 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Qwen3.6原生视觉语言系列Flash模型,模型效果相较3.5-Flash显著提升。本模型重点提升agentic coding能力(在多项代码智能体基准上大幅超越前代)、数学推理和代码推理能力;视觉方面在空间智能能力上显著增强,物体定位与目标检测提升尤为突出。

Starting at $0.17 / 1M tokensInput$0.99 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Qwen3.6-Max-Preview is a proprietary frontier model from Alibaba Cloud built on a sparse mixture-of-experts architecture with approximately 1 trillion total parameters. It is optimized for agentic coding, tool use, and...

Starting at $1.24 / 1M tokensInput$7.43 / 1M tokensOutput256kContext
Best forGeneral chat via Qwen, API workloads
RoutingConfigured
StreamingTool callingJSON modeLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

OpenAI 2026 主力旗舰,1M 上下文、原生自主智能体,可独立完成编码、研究、数据分析、跨工具办公全流程,速度与前代持平、智能大幅跃升

Starting at $5 / 1M tokensInput$30 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
DeepSeekCatalog

旗舰级 MoE 大模型,总参1.6T、激活 49B,原生支持百万级超长上下文。依托海量高质量训练数据,具备顶尖数学逻辑、复杂推理、专业代码与长文本深度解析能力,适配高阶科研、复杂办公、深度智能代理等高难度场景。

$1.65 / 1M tokensInput$3.31 / 1M tokensOutput1MContext
Best forChinese Q&A, general chat
RoutingConfigured
StreamingTool callingJSON modeLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
DeepSeekCatalog

高效轻量化MoE模型,总参284B,激活13B,原生支持百万超长上下文能力。推理速度快、延迟低、调用成本低廉,综合能力均衡,主打高并发、轻量化任务,适合日常对话、内容创作、基础 RAG、批量文案处理等普惠刚需场景。

$0.14 / 1M tokensInput$0.28 / 1M tokensOutput1MContext
Best forChinese Q&A, general chat
RoutingConfigured
StreamingTool callingJSON modeLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
MoonshotaiCatalog

kimi-k2.6是Kimi最新最智能的模型,具备更强更稳的长程代码编写能力,指令遵循和自我纠错能力显著提升,同时支持文本、图片与视频输入,思考与非思考模式,对话与Agent任务。

$0.9 / 1M tokensInput$3.72 / 1M tokensOutput256kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
Z AiCatalog

Available through the NextModel gateway via Z Ai.

$1.40 / 1M tokensInput$4.40 / 1M tokensOutputContext
Best forGeneral chat via Z Ai, API workloads
RoutingDegraded
Streaming
Platform curatedNextModel gateway catalog (Go origin)
View details
Z AiCatalog

GLM-5.1是智谱AI推出的面向长程任务(Long Horizon Task)设计的模型,总参数744B,支持200K超长上下文,最大输出 128K tokens。拥有强大逻辑推理、长文本理解与代码生成能力、兼顾性能与推理效率;在多任务基准中表现优异,适用于智能交互、企业应用、开发辅助等场景。

Starting at $0.83 / 1M tokensInput$3.31 / 1M tokensOutput202kContext
Best forGeneral chat via Z Ai, API workloads
RoutingConfigured
StreamingTool callingJSON modeLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Qwen3.6原生视觉语言系列Plus模型,展现出与当前顶尖前沿模型相媲美的卓越性能,模型效果相较3.5系列显著提升。模型在Agentic coding、前端编程、Vibe coding等代码能力、多模态万物识别、OCR、物体定位等能力上显著增强。

Starting at $0.28 / 1M tokensInput$1.66 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
GoogleCatalog

Gemini 3.7 Flash 是 Gemini 3 系列中兼顾高效性能与高性价比的主力模型。它具备媲美专业级的智能自主代理能力,代码生成与终端执行能力实现跨越式提升。

$1.50 / 1M tokensInput$7.50 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingDegraded
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT‑5.4 Nano 是 GPT‑5.4 最轻量、最快速的版本,专为对速度和成本要求极高的任务而设计。

$0.2 / 1M tokensInput$1.25 / 1M tokensOutput400kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-5.4 Mini将GPT-5.4的优势融入到一个更快、更高效的模型中,专为高负载工作量设计。

$0.75 / 1M tokensInput$4.50 / 1M tokensOutput400kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-5.4 是 OpenAI 2026年3月发布的旗舰推理模型,核心新特性包括:105万 Token 超长上下文窗口(输出支持12.8万 Token)、原生计算机使用能力(自主操作浏览器与软件)、多档推理强度调节(none 到 xhigh)、多模态理解(文本+图像)

Starting at $2.50 / 1M tokensInput$15 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-5.4 Pro使用更多计算资源来更深入地思考并提供始终更好的答案。仅可通过响应API访问,以在响应API请求前支持多轮模型交互功能,以及未来其他高级API特性。

Starting at $30 / 1M tokensInput$180 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-5.3-Codex is OpenAI’s most advanced agentic coding model, combining the frontier software engineering performance of GPT-5.2-Codex with the broader reasoning and professional knowledge capabilities of GPT-5.2. It achieves state-of-the-art results...

$1.75 / 1M tokensInput$14 / 1M tokensOutput400kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
AnthropicCatalog

Anthropic Claude 4.6 Sonnet,接近 Opus 级别的能力但成本降低约 40%。支持自适应思考(Adaptive Thinking),1M token 上下文窗口(Beta),在编码、计算机操作、长上下文推理和智能体规划方面全面提升。

$3 / 1M tokensInput$15 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
Z AiCatalog

GLM-5是面向Coding与Agent场景的新一代大模型,在复杂系统工程与长程任务中达到开源 SOTA,真实编程体验逼近 Claude Opus 级别;基于 744B 新基座、异步强化学习与稀疏注意力,实现从“写代码”到“写工程”的全面升级。

Starting at $0.58 / 1M tokensInput$2.58 / 1M tokensOutput198kContext
Best forGeneral chat via Z Ai, API workloads
RoutingConfigured
StreamingTool callingJSON modeLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
MiniMaxCatalog

MiniMax-M2.5是MiniMax推出的旗舰级开源大模型,经过数十万个真实复杂环境中的大规模强化学习训练,M2.5 在编程、工具调用和搜索、办公等生产力场景都达到或者刷新了行业的 SOTA。

$0.31 / 1M tokensInput$1.22 / 1M tokensOutput200kContext
Best forChinese Q&A, general chat
RoutingConfigured
StreamingTool callingJSON modeLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
AnthropicCatalog

Opus 4.6 is Anthropic’s strongest model for coding and long-running professional tasks. It is built for agents that operate across entire workflows rather than single prompts, making it especially effective...

$5 / 1M tokensInput$25 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
MoonshotaiCatalog

Kimi K2.5是2026年1月27日由月之暗面Kimi发布的新一代开源模型。该模型基于原生多模态架构设计,支持视觉与文本输入,集成了视觉理解与推理、编程、智能体等多种能力。

$0.58 / 1M tokensInput$3.02 / 1M tokensOutput256kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-5.2 Codex 优化版,专为智能体编码工作流设计。支持上下文压缩,在 SWE-Bench Pro (56.4%) 和 Terminal-Bench 2.0 (64.0%) 上达到 SOTA。

$1.75 / 1M tokensInput$14 / 1M tokensOutput400kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
AnthropicCatalog

Anthropic Claude 4.5 Haiku,极速低成本模型。性能接近 Claude Sonnet 4,适合大规模部署、多智能体协作和对延迟敏感的场景。支持视觉理解和函数调用。

$1 / 1M tokensInput$5 / 1M tokensOutput200kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-4o 的低成本精简版,兼顾基础多模态能力与经济性,适合大规模轻量级应用OpenAI。

$0.15 / 1M tokensInput$0.6 / 1M tokensOutput128kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-5.2 推理模型,引入响应压缩和 xhigh 推理深度。三层智能系统(Instant/Thinking/Pro)自动优化响应质量和速度,在专业知识工作基准测试中超越行业专家。

$1.75 / 1M tokensInput$14 / 1M tokensOutput400kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-5.1 Codex 增强版,OpenAI 前沿智能体编码模型。原生支持跨多个上下文窗口的压缩操作,可在单个任务中处理数百万 token,适合多小时持续编码会话和大型项目重构。

$1.25 / 1M tokensInput$10 / 1M tokensOutput400kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
AnthropicCatalog

Anthropic Claude 4.5 Opus,强大推理能力。支持最高 1M token 上下文窗口(特殊模式),擅长处理整个代码库、长文档和多日对话历史。扩展思考模式下可进行深度推理。

$5 / 1M tokensInput$25 / 1M tokensOutput200kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-5.1 Codex 轻量版,平衡编码能力与成本效率。适合日常开发辅助、代码补全和快速迭代场景。

$0.25 / 1M tokensInput$2 / 1M tokensOutput400kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-5.1 Codex 优化版,在 GPT-5 Codex 基础上提升代码生成质量和多文件重构能力。支持上下文压缩,适合长时间编码会话。

$1.25 / 1M tokensInput$10 / 1M tokensOutput400kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-5.1 推理模型,在 GPT-5 基础上提升智能水平和指令遵循能力。更温暖自然的对话风格,简单任务更快响应,复杂任务更持久推理。

$1.25 / 1M tokensInput$10 / 1M tokensOutput400kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-5 专业推理模型,提供最高级别的推理深度和准确性。适合需要极致推理能力的科研、金融分析和复杂决策场景,支持更长的思考时间。

$15 / 1M tokensInput$120 / 1M tokensOutput400kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
GoogleCatalog

Google Gemini 3.1 Pro,最新旗舰推理模型。在 Gemini 3 Pro 基础上全面提升,1M token 上下文窗口,支持文本、图像、视频、音频多模态输入。ARC-AGI-2 得分 77.1%,编码和推理能力大幅增强。

Starting at $2 / 1M tokensInput$12 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingDegraded
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
AnthropicCatalog

Anthropic Claude 4.5 Sonnet,平衡性能与成本的中端模型。200K 上下文窗口,在编码、分析和智能体推理方面表现出色,支持扩展思考模式。

$3 / 1M tokensInput$15 / 1M tokensOutput200kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-5 推理模型,OpenAI 第五代旗舰。统一快速响应与深度思考模式,400K 总上下文(272K 输入 + 128K 输出)。在编码、数学、写作、健康和视觉感知等领域达到 SOTA 水平。

$1.25 / 1M tokensInput$10 / 1M tokensOutput400kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-5 轻量推理模型,平衡性能与成本。继承 GPT-5 核心推理能力,适合高吞吐量批处理和对延迟敏感的场景。

$0.25 / 1M tokensInput$2 / 1M tokensOutput400kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-5 超轻量推理模型,极低成本和延迟。适合简单查询、分类任务和大规模并发调用场景。

$0.05 / 1M tokensInput$0.4 / 1M tokensOutput400kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

O4 轻量推理模型,继承 O3 的工具调用和多模态推理能力,成本更低、速度更快。适合日常推理任务和对延迟敏感的应用。

$1.10 / 1M tokensInput$4.40 / 1M tokensOutput200kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-4.1 多模态模型,支持最高 1M token 上下文窗口。在编码、指令遵循和长上下文理解方面大幅提升,支持文本和图像输入,适用于大规模文档处理和复杂代码任务。

$2 / 1M tokensInput$8 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-4.1 轻量版,1M token 上下文窗口,在保持强大编码和推理能力的同时大幅降低成本。适合高吞吐量场景和对延迟敏感的应用。

$0.4 / 1M tokensInput$1.60 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

GPT-4.1 超轻量版,1M token 上下文窗口,极低成本和延迟。支持多模态输入,适合分类、自动补全、标签提取等轻量级任务。

$0.1 / 1M tokensInput$0.4 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

O3 增强推理模型,首次支持智能体式工具组合调用。可自主搜索网页、分析文件、深度推理视觉输入并生成图像,在数学竞赛 AIME 中达到 95.2% 准确率。

$2 / 1M tokensInput$8 / 1M tokensOutput200kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

OpenAI 旗舰多模态模型,支持文本、图像和音频输入输出。128K 上下文窗口,具备出色的指令遵循和多语言能力,适用于复杂对话、内容生成和分析任务。

$2.50 / 1M tokensInput$10 / 1M tokensOutput128kContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingTool callingJSON modeVisionLong context
Platform curatedNextModel gateway catalog (Go origin)
View details
DoubaoCatalog

字节跳动旗舰音视频联合生成模型,支持文本 / 图像 / 音频 / 视频多模态输入,具备原生多镜头叙事与角色一致性,适配专业级视频创作与编辑。官方API参考:

$0.148 / sPer secondContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
DoubaoCatalog

极速版视频生成模型,推理速度提升 10 倍 +,30% 更快生成且不牺牲质量,支持多镜头自动编排与原生音频同步,适配快速内容生产场景。官方API参考:

$0.119 / sPer secondContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
DoubaoCatalog

Seedance 2.0 mini 是面向更广泛视频生成需求推出的新一代高性价比视频生成模型。在保持竞争力效果的同时,将视频生成能力带入更低门槛、更高频、更规模化的应用场景

$0.0738 / sPer secondContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
DreaminaCatalog

旗舰音视频联合生成模型,支持文本 / 图像 / 音频 / 视频多模态输入,具备原生多镜头叙事与角色一致性,适配专业级视频创作与编辑。官方API参考:

$0.153 / sPer secondContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
DreaminaCatalog

极速版视频生成模型,推理速度提升 10 倍 +,30% 更快生成且不牺牲质量,支持多镜头自动编排与原生音频同步,适配快速内容生产场景。官方API参考:

$0.122 / sPer secondContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
DreaminaCatalog

Doubao-Seedance-2.0-mini 是豆包大模型团队推出的高性价比视频生成模型,面向更广泛的视频创作与生产需求。相较 Seedance 2.0,2.0 mini 在价格上进一步下探,生成成本降低约 50%,帮助企业客户以更低成本完成调用。2.0 mini 兼顾成本与可用性,适用于电商内容生产、营销素材批量生成、UGC 内容创作、特效玩法生成等高频、规模化的视频生成场景。官方API参考:

$0.0764 / sPer secondContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
GeminiCatalog

谷歌旗舰专业图像生成模型,拥有工作室级精细画面控制、顶尖文字排版与强现实世界认知,4K 高清输出,面向专业设计、高精度可视化与复杂商业创作需求。

$0.314 / imagePer imageContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
GeminiCatalog

谷歌高速高性价比图像生成模型,兼顾 Pro 级画面质量与极速推理,文字渲染精准、支持 4K 输出与检索增强生成,适配大批量、实时图像创作场景。

$0.151 / imagePer imageContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
GeminiCatalog

Gemini 3.1 Flash Lite Image(Nano Banana 2 Lite)模型是图片生成系列中的效率专家,专为超低延迟和经济高效的图片生成和修改而设计。

$0.0336 / imagePer imageContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
KlingCatalog

支持文本、图像及多图参考输入,在文本响应度与画面美感上实现全面升级,提供标准与高品质双模式,适用于多角色交互与复杂场景的视频创作。【积分单价说明:每1积分单价为0.15USD】

$0.525 / sPer secondContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
KlingCatalog

支持文本与图像输入,在生成稳定性与创意想象力上全面升级,兼顾高性能与高性价比,适用于大批量视频生成与企业级内容生产。【积分单价说明:每1积分单价为0.15USD】

$0.075 / sPer secondContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
KlingCatalog

支持画面与语音、音效、环境音同步生成,运动控制能力增强,适用于有声广告、短视频与多媒体内容创作。【积分单价说明:每1积分单价为0.15USD】

$0.18 / sPer secondContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
KlingCatalog

支持智能分镜与15秒长视频生成,最高4K Ultra画质输出,可实现场景切换与连续叙事,适用于企业广告营销与专业影视创作。【积分单价说明:每1积分单价为0.15USD】

$0.45 / sPer secondContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

Available through the NextModel gateway via OpenAI.

$0.15 / 10k charsInputOutputContext
Best forGeneral chat via OpenAI, API workloads
RoutingConfigured
Platform curatedNextModel gateway catalog (Go origin)
View details
OpenAICatalog

Available through the NextModel gateway via OpenAI.

$0.3 / 10k charsInputOutputContext
Best forGeneral chat via OpenAI, API workloads
RoutingConfigured
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Qwen-Image-2.0系列加速版模型,实现了图片生成和图片编辑的融合;具备更专业的文字渲染1k token指令支持能力、更细腻的真实质感,细腻刻画写实场景、更强的语义遵循能力。加速版有效实现了模型效果和性能的最佳平衡。官方API参考:

$0.0287 / imagePer imageContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Qwen-Image-2.0系列加速版模型,实现了图片生成和图片编辑的融合;具备更专业的文字渲染1k token指令支持能力、更细腻的真实质感,细腻刻画写实场景、更强的语义遵循能力。加速版有效实现了模型效果和性能的最佳平衡。官方API参考:

$0.035 / imagePer imageContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Available through the NextModel gateway via Qwen.

$0.0717 / imagePer imageContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

Available through the NextModel gateway via Qwen.

$0.075 / imagePer imageContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

万相2.6-文生图,画面质感、美学表现、指令遵循升级,在艺术风格精准控制、真实感人像、长文本生图及广泛历史文化IP覆盖上均表现出卓越能力,可生成高质量且富有表现力的视觉内容。 官方API网址:

$0.0287 / imagePer imageContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

万相2.6-文生图,画面质感、美学表现、指令遵循升级,在艺术风格精准控制、真实感人像、长文本生图及广泛历史文化IP覆盖上均表现出卓越能力,可生成高质量且富有表现力的视觉内容。 官方API文档参考:

$0.03 / imagePer imageContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

万相2.7-图生视频,演绎能力全面升级,文戏情感细腻自然,动作戏激烈拳拳到肉,搭配更富有戏剧性和节奏感的镜头切换,实现更强表演能力。官方API参考:

$0.138 / sPer secondContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

万相2.7-图生视频,演绎能力全面升级,文戏情感细腻自然,动作戏激烈拳拳到肉,搭配更富有戏剧性和节奏感的镜头切换,实现更强表演能力。官方API参考:

$0.15 / sPer secondContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

万相2.7-图像生成与编辑,支持文生图、文生组图、图生组图、图像编辑、多图参考生成、交互式编辑,在文字渲染、主体一致性、复杂指令遵循上都有更强表现。官方API参考:

$0.0275 / imagePer imageContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

万相2.7-图像生成与编辑,支持文生图、文生组图、图生组图、图像编辑、多图参考生成、交互式编辑,在文字渲染、主体一致性、复杂指令遵循上都有更强表现。官方API参考:

$0.03 / imagePer imageContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

万相2.7-图像生成与编辑旗舰版模型,支持文生图、文生组图、图生组图、图像编辑、多图参考生成、交互式编辑,在文字渲染、主体一致性、复杂指令遵循上都有更强表现。官方API参考:

$0.0688 / imagePer imageContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

万相2.7-图像生成与编辑旗舰版模型,支持文生图、文生组图、图生组图、图像编辑、多图参考生成、交互式编辑,在文字渲染、主体一致性、复杂指令遵循上都有更强表现。官方API参考:

$0.075 / imagePer imageContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

万相2.7-参考生视频,更加稳定的角色、道具与场景参考,支持最大5个图/视频混合参考,支持音频音色参考,搭配基础能力升级实现更强表演能力。官方API参考:

$0.138 / sPer secondContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

万相2.7-参考生视频,更加稳定的角色、道具与场景参考,支持最大5个图/视频混合参考,支持音频音色参考,搭配基础能力升级实现更强表演能力。官方API参考:

$0.15 / sPer secondContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

万相2.7-文生视频,演绎能力全面升级,文戏情感细腻自然,动作戏激烈拳拳到肉,搭配更富有戏剧性和节奏感的镜头切换,实现更强表演能力。官方API参考:

$0.138 / sPer secondContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
QwenCatalog

万相2.7-文生视频,演绎能力全面升级,文戏情感细腻自然,动作戏激烈拳拳到肉,搭配更富有戏剧性和节奏感的镜头切换,实现更强表演能力。官方API参考:

$0.15 / sPer secondContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
ViduCatalog

Vidu Q3 Pro 是生数科技推出的高品质视频生成模型,支持文生视频、图生视频和首尾帧视频三种工作流,可生成带同步音频的 1080p 高清视频,最长达 16 秒,具备极致的角色保真度、运动一致性和风格控制能力。【积分单价说明:每1积分单价为0.046USD】

$0.11 / sPer secondContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details
ViduCatalog

Vidu Q3 Turbo 是 Vidu Q3 系列的快速版 AI 视频生成模型,能将文本提示、静态图像或首尾帧对快速转换为流畅的高清视频并带有同步音频,在保持高质量输出的同时大幅提升生成速度,兼顾效率与成本。【积分单价说明:每1积分单价为0.046USD】

$0.0598 / sPer secondContext
Best forimage understanding, multimodal chat
RoutingConfigured
StreamingVision
Platform curatedNextModel gateway catalog (Go origin)
View details

18 video models

Video generation models

vapeurcertified

kling/kling-v1-6-cn

$0.525 / soutput video
T2VI2V first frameI2V first + lastR2V references
Durations5s
Resolutions720p
References1-4 reference images
vapeurcertified

kling/kling-v2-6-cn

$0.18 / soutput video
T2VI2V first frameI2V first + last
Durations5s
Resolutions720p
vapeurcertified

kling/kling-v3-cn

$0.45 / soutput video
T2VI2V first frameI2V first + last
Durations5s
Resolutions720p
vapeurcertified

qwen/wan2.7-i2v-cn

$0.138 / soutput video
I2V first frameI2V first + last
Durations5s
Resolutions720p, 1080p
vapeurcertified

qwen/wan2.7-i2v-glb

$0.15 / soutput video
I2V first frameI2V first + last
Durations5s
Resolutions720p, 1080p
vapeurcertified

qwen/wan2.7-r2v-cn

$0.138 / soutput video
R2V references
Durations5s
Resolutions720p, 1080p
References1-4 reference images
vapeurcertified

qwen/wan2.7-r2v-glb

$0.15 / soutput video
R2V references
Durations5s
Resolutions720p, 1080p
References1-4 reference images
vapeurcertified

qwen/wan2.7-t2v-cn

$0.138 / soutput video
T2V
Durations5s
Resolutions720p, 1080p
vapeurcertified

vidu/viduq3-pro-cn

$0.11 / soutput video
T2VI2V first frameI2V first + last
Durations5s
Resolutions540p, 720p, 1080p
vapeurcertified

vidu/viduq3-turbo-cn

$0.0598 / soutput video
T2VI2V first frameI2V first + last
Durations5s
Resolutions540p, 720p, 1080p
ModelProviderInputOutputContextCapabilitiesBest forLatencyStatusSource
Claude Fable 5.1anthropic/claude-fable-5.1Anthropic$10 / 1M tokens$50 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Qwen3.8-Max-GLBqwen/qwen3.8-max-glbQwen$2 / 1M tokens$6 / 1M tokens1M
Streaming
General chat via Qwen, API workloads1957-6020msCatalogPlatform curated
FLUX.2 Flexblack-forest-labs/FLUX.2-flexBlack Forest Labs$0.2 / image
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
FLUX.2 PROblack-forest-labs/FLUX.2-proBlack Forest Labs$0.075 / image
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
GPT Image 1openai/gpt-image-1OpenAI$0.26 / image
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
GPT Image 1 Miniopenai/gpt-image-1-miniOpenAI$0.0539 / image
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
GPT Image 1.5openai/gpt-image-1.5OpenAI$0.218 / image
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
GPT Image 2.0openai/gpt-image-2OpenAI$0.221 / image
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Text Embedding 3 Smallopenai/text-embedding-3-smallOpenAI$0.02 / 1M tokens$0 / 1M tokens
Streaming
General chat via OpenAI, API workloads0-0msCatalogPlatform curated
Text Embedding 3 Largeopenai/text-embedding-3-largeOpenAI$0.13 / 1M tokens$0 / 1M tokens
Streaming
General chat via OpenAI, API workloads0-0msCatalogPlatform curated
GLM 5.3z-ai/glm-5.3Z Ai$1.40 / 1M tokens$4.40 / 1M tokens1M
StreamingTool callingJSON modeLong context
General chat via Z Ai, API workloads0-0msCatalogPlatform curated
GPT-5 Codexopenai/gpt-5-codexOpenAI$1.25 / 1M tokens$10 / 1M tokens400k
Streaming
General chat via OpenAI, API workloads845-3461msCatalogPlatform curated
Qwen3-VL-Flash CNqwen/qwen3-vl-flash-cnQwenStarting at $0.03 / 1M tokens$0.22 / 1M tokens256k
Streaming
General chat via Qwen, API workloads835-1062msCatalogPlatform curated
Qwen3.5 Flashqwen/qwen3.5-flashQwenStarting at $0.03 / 1M tokens$0.29 / 1M tokens1M
Streaming
General chat via Qwen, API workloads719-1339msCatalogPlatform curated
Qwen3-VL-Flash GLBqwen/qwen3-vl-flash-glbQwenStarting at $0.05 / 1M tokens$0.4 / 1M tokens256k
Streaming
General chat via Qwen, API workloads613-929msCatalogPlatform curated
Qwen3.5 Plusqwen/qwen3.5-plusQwenStarting at $0.12 / 1M tokens$0.69 / 1M tokens1M
Streaming
General chat via Qwen, API workloads2574-3122msCatalogPlatform curated
Qwen3-VL-Plus CNqwen/qwen3-vl-plus-cnQwenStarting at $0.15 / 1M tokens$1.44 / 1M tokens256k
Streaming
General chat via Qwen, API workloads976-1249msCatalogPlatform curated
Qwen3.6 Flash GLBqwen/qwen3.6-flash-glbQwenStarting at $0.25 / 1M tokens$1.50 / 1M tokens1M
Streaming
General chat via Qwen, API workloads3566-3725msCatalogPlatform curated
Qwen3-Max CNqwen/qwen3-max-cnQwenStarting at $0.36 / 1M tokens$1.44 / 1M tokens256k
Streaming
General chat via Qwen, API workloads1219-1427msCatalogPlatform curated
Qwen3-VL-Plus GLBqwen/qwen3-vl-plus-2025-12-19-glbQwenStarting at $0.2 / 1M tokens$1.60 / 1M tokens256k
Streaming
General chat via Qwen, API workloads570-1200msCatalogPlatform curated
Qwen3.6 Plus GLBqwen/qwen3.6-plus-glbQwenStarting at $0.5 / 1M tokens$3 / 1M tokens1M
Streaming
General chat via Qwen, API workloads5535-6156msCatalogPlatform curated
Qwen3-Max GLBqwen/qwen3-max-2026-01-23-glbQwenStarting at $1.20 / 1M tokens$6 / 1M tokens256k
Streaming
General chat via Qwen, API workloads1036-1620msCatalogPlatform curated
Qwen3.6 Max GLBqwen/qwen3.6-max-preview-glbQwenStarting at $1.30 / 1M tokens$7.80 / 1M tokens256k
Streaming
General chat via Qwen, API workloads7045-9820msCatalogPlatform curated
Deepseek-V4-Pro-0813deepseek/deepseek-v4-pro-0813DeepSeek$1.33 / 1M tokens$3.98 / 1M tokens1M
StreamingTool callingJSON modeLong context
Chinese Q&A, general chat1668-14930msCatalogPlatform curated
Doubao Seed 2.0 Minidoubao-seed-2-0-miniVolcengineStarting at $0.03 / 1M tokens$0.3 / 1M tokens256k
Streaming
Chinese Q&A, general chat2020-3200msCatalogPlatform curated
Doubao Seed 2.0 Prodoubao-seed-2-0-proVolcengineStarting at $0.48 / 1M tokens$2.36 / 1M tokens256k
Streaming
Chinese Q&A, general chat6244-7454msCatalogPlatform curated
Doubao Seed 2.0 Litedoubao-seed-2-0-liteVolcengineStarting at $0.09 / 1M tokens$0.53 / 1M tokens256k
Streaming
Chinese Q&A, general chat20776-27806msCatalogPlatform curated
Doubao Seed 2.0 Codedoubao-seed-2-0-codeVolcengineStarting at $0.48 / 1M tokens$2.36 / 1M tokens256k
Streaming
Chinese Q&A, general chat3588-4297msCatalogPlatform curated
Qwen3.5 Flash GLBqwen/qwen3.5-flash-glbQwen$0.1 / 1M tokens$0.4 / 1M tokens1M
Streaming
General chat via Qwen, API workloads1628-3840msCatalogPlatform curated
grok-4-1-fast-non-reasoningx-ai/grok-4.1-fast-non-reasoningX Ai$0.2 / 1M tokens$0.5 / 1M tokens2M
Streaming
General chat via X Ai, API workloads876-2590msCatalogPlatform curated
grok-4-20-non-reasoningx-ai/grok-4.20-non-reasoningX Ai$1.25 / 1M tokens$2.50 / 1M tokens256k
Streaming
General chat via X Ai, API workloads825-2560msCatalogPlatform curated
grok-4-20-reasoningx-ai/grok-4.20-reasoningX Ai$1.25 / 1M tokens$2.50 / 1M tokens256k
Streaming
General chat via X Ai, API workloads1970-3116msCatalogPlatform curated
Doubao-Seed-2.1-prodoubao-seed-2-1-proVolcengine$0.89 / 1M tokens$4.42 / 1M tokens256k
Streaming
Chinese Q&A, general chat12630-20216msCatalogPlatform curated
Qwen3.7 Max GLBqwen/qwen3.7-max-glbQwen$2.50 / 1M tokens$7.50 / 1M tokens1M
Streaming
General chat via Qwen, API workloads6212-9866msCatalogPlatform curated
Doubao-Seed-2.1-turbodoubao-seed-2-1-turboVolcengine$0.45 / 1M tokens$2.21 / 1M tokens256k
Streaming
Chinese Q&A, general chat9419-24588msCatalogPlatform curated
grok-4-1-fast-reasoningx-ai/grok-4.1-fast-reasoningX Ai$0.2 / 1M tokens$0.5 / 1M tokens2M
Streaming
General chat via X Ai, API workloads2802-28567msCatalogPlatform curated
Qwen3.8-Maxqwen/qwen3.8-maxQwen$1.77 / 1M tokens$5.30 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat1292-1672msCatalogPlatform curated
DeepSeek-V4-Flash-0731deepseek/deepseek-v4-flash-0731DeepSeek$0.45 / 1M tokens$1.33 / 1M tokens1M
StreamingTool callingJSON modeLong context
Chinese Q&A, general chat1728-3730msCatalogPlatform curated
Qwen3.7 Flashqwen/qwen3.7-flashQwenStarting at $0.03 / 1M tokens$0.12 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat939-1189msCatalogPlatform curated
Gemini 3.5 Flash-Litegoogle/gemini-3.5-flash-liteGoogle$0.3 / 1M tokens$2.50 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat923-1223msCatalogPlatform curated
Gemini 3.8 Flashgoogle/gemini-3.8-flashGoogle$0.75 / 1M tokens$3.75 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat0-0msCatalogPlatform curated
GPT-5.6 Lunaopenai/gpt-5.6-lunaOpenAIStarting at $0.2 / 1M tokens$1.20 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat1791-3004msCatalogPlatform curated
GPT-5.6 Terraopenai/gpt-5.6-terraOpenAIStarting at $2 / 1M tokens$12 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat1754-3372msCatalogPlatform curated
GPT-5.6 Solopenai/gpt-5.6-solOpenAIStarting at $5 / 1M tokens$30 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat1911-3000msCatalogPlatform curated
Qwen3.5 Plus GLBqwen/qwen3.5-plus-glbQwenStarting at $0.4 / 1M tokens$2.40 / 1M tokens1M
Streaming
General chat via Qwen, API workloads9803-29361msCatalogPlatform curated
Kimi-K2.7-Codemoonshotai/kimi-k2.7-codeMoonshotai$0.9 / 1M tokens$3.72 / 1M tokens256k
StreamingTool callingJSON modeVision
image understanding, multimodal chat988-1072msCatalogPlatform curated
GLM-5.2z-ai/glm-5.2Z Ai$1.18 / 1M tokens$4.12 / 1M tokens1M
StreamingTool callingJSON modeLong context
General chat via Z Ai, API workloads1103-2070msCatalogPlatform curated
Qwen3.7 Plusqwen/qwen3.7-plusQwenStarting at $0.3 / 1M tokens$1.18 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat1204-1480msCatalogPlatform curated
Qwen3.7 Maxqwen/qwen3.7-maxQwen$1.65 / 1M tokens$4.96 / 1M tokens1M
StreamingTool callingJSON modeLong context
General chat via Qwen, API workloads1125-1692msCatalogPlatform curated
Gemini 3.1 Flash-Litegoogle/gemini-3.1-flash-liteGoogle$0.25 / 1M tokens$1.50 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat996-5185msCatalogPlatform curated
GPT-Chat Latestopenai/gpt-chat-latestOpenAI$5 / 1M tokens$30 / 1M tokens400k
StreamingTool callingJSON modeVision
image understanding, multimodal chat2075-2747msCatalogPlatform curated
Grok 4.3x-ai/grok-4.3X Ai$1.25 / 1M tokens$2.50 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat2516-4795msCatalogPlatform curated
Qwen3.6 Flashqwen/qwen3.6-flashQwenStarting at $0.17 / 1M tokens$0.99 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat578-686msCatalogPlatform curated
Qwen3.6 Maxqwen/qwen3.6-max-previewQwenStarting at $1.24 / 1M tokens$7.43 / 1M tokens256k
StreamingTool callingJSON modeLong context
General chat via Qwen, API workloads7028-11803msCatalogPlatform curated
GPT-5.5openai/gpt-5.5OpenAIStarting at $5 / 1M tokens$30 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat1724-2781msCatalogPlatform curated
DeepSeek-V4-Prodeepseek/deepseek-v4-proDeepSeek$1.65 / 1M tokens$3.31 / 1M tokens1M
StreamingTool callingJSON modeLong context
Chinese Q&A, general chat1477-1902msCatalogPlatform curated
DeepSeek-V4-Flashdeepseek/deepseek-v4-flashDeepSeek$0.14 / 1M tokens$0.28 / 1M tokens1M
StreamingTool callingJSON modeLong context
Chinese Q&A, general chat1548-6515msCatalogPlatform curated
Kimi-K2.6moonshotai/kimi-k2.6Moonshotai$0.9 / 1M tokens$3.72 / 1M tokens256k
StreamingTool callingJSON modeVision
image understanding, multimodal chat800-1433msCatalogPlatform curated
GLM 5.3 CNz-ai/glm-5.3-cnZ Ai$1.40 / 1M tokens$4.40 / 1M tokens
Streaming
General chat via Z Ai, API workloads0-0msCatalogPlatform curated
GLM-5.1z-ai/glm-5.1Z AiStarting at $0.83 / 1M tokens$3.31 / 1M tokens202k
StreamingTool callingJSON modeLong context
General chat via Z Ai, API workloads1257-1988msCatalogPlatform curated
Qwen3.6 Plusqwen/qwen3.6-plusQwenStarting at $0.28 / 1M tokens$1.66 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat911-17203msCatalogPlatform curated
Gemini 3.7 Flashgoogle/gemini-3.7-flashGoogle$1.50 / 1M tokens$7.50 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat0-0msCatalogPlatform curated
GPT-5.4 Nanoopenai/gpt-5.4-nanoOpenAI$0.2 / 1M tokens$1.25 / 1M tokens400k
StreamingTool callingJSON modeVision
image understanding, multimodal chat1755-2235msCatalogPlatform curated
GPT-5.4 Miniopenai/gpt-5.4-miniOpenAI$0.75 / 1M tokens$4.50 / 1M tokens400k
StreamingTool callingJSON modeVision
image understanding, multimodal chat1459-2863msCatalogPlatform curated
GPT-5.4openai/gpt-5.4OpenAIStarting at $2.50 / 1M tokens$15 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat1727-3122msCatalogPlatform curated
GPT-5.4 Proopenai/gpt-5.4-proOpenAIStarting at $30 / 1M tokens$180 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat7765-13899msCatalogPlatform curated
GPT-5.3 Codexopenai/gpt-5.3-codexOpenAI$1.75 / 1M tokens$14 / 1M tokens400k
StreamingTool callingJSON modeVision
image understanding, multimodal chat2440-3122msCatalogPlatform curated
Claude Sonnet 4.6anthropic/claude-sonnet-4.6Anthropic$3 / 1M tokens$15 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat1745-2526msCatalogPlatform curated
GLM-5z-ai/glm-5Z AiStarting at $0.58 / 1M tokens$2.58 / 1M tokens198k
StreamingTool callingJSON modeLong context
General chat via Z Ai, API workloads1293-1662msCatalogPlatform curated
MiniMax M2.5minimax/minimax-m2.5MiniMax$0.31 / 1M tokens$1.22 / 1M tokens200k
StreamingTool callingJSON modeLong context
Chinese Q&A, general chat957-1303msCatalogPlatform curated
Claude Opus 4.6anthropic/claude-opus-4.6Anthropic$5 / 1M tokens$25 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat2264-2919msCatalogPlatform curated
Kimi-K2.5moonshotai/kimi-k2.5Moonshotai$0.58 / 1M tokens$3.02 / 1M tokens256k
StreamingTool callingJSON modeVision
image understanding, multimodal chat1119-1742msCatalogPlatform curated
GPT-5.2 Codexopenai/gpt-5.2-codexOpenAI$1.75 / 1M tokens$14 / 1M tokens400k
StreamingTool callingJSON modeVision
image understanding, multimodal chat2151-3959msCatalogPlatform curated
Claude Haiku 4.5anthropic/claude-haiku-4.5Anthropic$1 / 1M tokens$5 / 1M tokens200k
StreamingTool callingJSON modeVision
image understanding, multimodal chat1178-3573msCatalogPlatform curated
GPT-4o-miniopenai/gpt-4o-miniOpenAI$0.15 / 1M tokens$0.6 / 1M tokens128k
StreamingTool callingJSON modeVision
image understanding, multimodal chat1505-2808msCatalogPlatform curated
GPT-5.2openai/gpt-5.2OpenAI$1.75 / 1M tokens$14 / 1M tokens400k
StreamingTool callingJSON modeVision
image understanding, multimodal chat2177-3348msCatalogPlatform curated
GPT-5.1 Codex Maxopenai/gpt-5.1-codex-maxOpenAI$1.25 / 1M tokens$10 / 1M tokens400k
StreamingTool callingJSON modeVision
image understanding, multimodal chat1232-3138msCatalogPlatform curated
Claude Opus 4.5anthropic/claude-opus-4.5Anthropic$5 / 1M tokens$25 / 1M tokens200k
StreamingTool callingJSON modeVision
image understanding, multimodal chat1949-4380msCatalogPlatform curated
GPT-5.1 Codex Miniopenai/gpt-5.1-codex-miniOpenAI$0.25 / 1M tokens$2 / 1M tokens400k
StreamingTool callingJSON modeVision
image understanding, multimodal chat1378-2489msCatalogPlatform curated
GPT-5.1 Codexopenai/gpt-5.1-codexOpenAI$1.25 / 1M tokens$10 / 1M tokens400k
StreamingTool callingJSON modeVision
image understanding, multimodal chat1850-2635msCatalogPlatform curated
GPT-5.1openai/gpt-5.1OpenAI$1.25 / 1M tokens$10 / 1M tokens400k
StreamingTool callingJSON modeVision
image understanding, multimodal chat2021-2542msCatalogPlatform curated
GPT-5 Proopenai/gpt-5-proOpenAI$15 / 1M tokens$120 / 1M tokens400k
StreamingTool callingJSON modeVision
image understanding, multimodal chat14726-19787msCatalogPlatform curated
Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-previewGoogleStarting at $2 / 1M tokens$12 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Claude Sonnet 4.5anthropic/claude-sonnet-4.5Anthropic$3 / 1M tokens$15 / 1M tokens200k
StreamingTool callingJSON modeVision
image understanding, multimodal chat2130-6376msCatalogPlatform curated
GPT-5openai/gpt-5OpenAI$1.25 / 1M tokens$10 / 1M tokens400k
StreamingTool callingJSON modeVision
image understanding, multimodal chat2810-5989msCatalogPlatform curated
GPT-5 Miniopenai/gpt-5-miniOpenAI$0.25 / 1M tokens$2 / 1M tokens400k
StreamingTool callingJSON modeVision
image understanding, multimodal chat1898-3267msCatalogPlatform curated
GPT-5 Nanoopenai/gpt-5-nanoOpenAI$0.05 / 1M tokens$0.4 / 1M tokens400k
StreamingTool callingJSON modeVision
image understanding, multimodal chat1869-3917msCatalogPlatform curated
O4 Miniopenai/o4-miniOpenAI$1.10 / 1M tokens$4.40 / 1M tokens200k
StreamingTool callingJSON modeVision
image understanding, multimodal chat1143-2415msCatalogPlatform curated
GPT-4.1openai/gpt-4.1OpenAI$2 / 1M tokens$8 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat2067-9317msCatalogPlatform curated
GPT-4.1 Miniopenai/gpt-4.1-miniOpenAI$0.4 / 1M tokens$1.60 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat1962-3592msCatalogPlatform curated
GPT-4.1 Nanoopenai/gpt-4.1-nanoOpenAI$0.1 / 1M tokens$0.4 / 1M tokens1M
StreamingTool callingJSON modeVision
image understanding, multimodal chat1413-18349msCatalogPlatform curated
O3openai/o3OpenAI$2 / 1M tokens$8 / 1M tokens200k
StreamingTool callingJSON modeVision
image understanding, multimodal chat2696-3772msCatalogPlatform curated
GPT-4oopenai/gpt-4oOpenAI$2.50 / 1M tokens$10 / 1M tokens128k
StreamingTool callingJSON modeVision
image understanding, multimodal chat1772-2933msCatalogPlatform curated
Doubao Seedance 2.0doubao/doubao-seedance-2-0-260128Doubao$0.148 / s
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Doubao Seedance 2.0 Fastdoubao/doubao-seedance-2-0-fast-260128Doubao$0.119 / s
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Doubao Seedance 2.0 minidoubao/doubao-seedance-2-0-mini-260615Doubao$0.0738 / s
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Dreamina-Seedance 2.0dreamina/dreamina-seedance-2-0-260128Dreamina$0.153 / s
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Dreamina-Seedance 2.0 Fastdreamina/dreamina-seedance-2-0-fast-260128Dreamina$0.122 / s
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Dreamina-Seedance 2.0 minidreamina/dreamina-seedance-2-0-mini-260615Dreamina$0.0764 / s
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Gemini 3 Pro Image (Nano Banana Pro) gemini/gemini-3-pro-imageGemini$0.314 / image
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Gemini 3.1 Flash Image (Nano Banana 2)gemini/gemini-3.1-flash-imageGemini$0.151 / image
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Gemini 3.1 Flash Lite Image(Nano Banana 2 Lite)gemini/gemini-3.1-flash-lite-imageGemini$0.0336 / image
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Kling V1.6 CNkling/kling-v1-6-cnKling$0.525 / s
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Kling V2.5 Turbo CNkling/kling-v2-5-turbo-cnKling$0.075 / s
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Kling V2.6 CNkling/kling-v2-6-cnKling$0.18 / s
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Kling V3 CNkling/kling-v3-cnKling$0.45 / s
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
TTS 1openai/tts-1OpenAI$0.15 / 10k chars
General chat via OpenAI, API workloads0-0msCatalogPlatform curated
TTS 1 HDopenai/tts-1-hdOpenAI$0.3 / 10k chars
General chat via OpenAI, API workloads0-0msCatalogPlatform curated
Qwen-Image-2.0 CNqwen/qwen-image-2.0-cnQwen$0.0287 / image
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Qwen-Image-2.0 GLBqwen/qwen-image-2.0-glbQwen$0.035 / image
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Qwen-Image-2.0-Pro CNqwen/qwen-image-2.0-pro-cnQwen$0.0717 / image
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Qwen-Image-2.0-Pro GLBqwen/qwen-image-2.0-pro-glbQwen$0.075 / image
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Wan2.6-T2I-CNqwen/wan2.6-t2i-cnQwen$0.0287 / image
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Wan2.6-T2I-GLBqwen/wan2.6-t2i-glbQwen$0.03 / image
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Wan2.7-I2V-CNqwen/wan2.7-i2v-cnQwen$0.138 / s
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Wan2.7-I2V-GLBqwen/wan2.7-i2v-glbQwen$0.15 / s
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Wan2.7 Image CNqwen/wan2.7-image-cnQwen$0.0275 / image
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Wan2.7 Image GLBqwen/wan2.7-image-glbQwen$0.03 / image
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Wan2.7 Image Pro CNqwen/wan2.7-image-pro-cnQwen$0.0688 / image
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Wan2.7 Image Pro GLBqwen/wan2.7-image-pro-glbQwen$0.075 / image
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Wan2.7-R2V-CNqwen/wan2.7-r2v-cnQwen$0.138 / s
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Wan2.7-R2V-GLBqwen/wan2.7-r2v-glbQwen$0.15 / s
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Wan2.7-T2V-CNqwen/wan2.7-t2v-cnQwen$0.138 / s
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Wan2.7-T2V-GLBqwen/wan2.7-t2v-glbQwen$0.15 / s
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Vidu-Q3-pro-CNvidu/viduq3-pro-cnVidu$0.11 / s
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated
Vidu-Q3-turbo-CNvidu/viduq3-turbo-cnVidu$0.0598 / s
StreamingVision
image understanding, multimodal chat0-0msCatalogPlatform curated