MODEL DIRECTORY
AI Model Directory
Context length, max output, modalities, reasoning support and official API pricing for Claude, GPT, Gemini, DeepSeek, Qwen and other major AI models, alongside the lowest AI gateway price. Updated continuously.
617Models
77Providers
138Relay sites
Updated 2026-11-05 08:15 (UTC+8)617 models foundPrices in USD per 1M tokens; relay prices and ≈ are converted from CNY at today's rate
| Model | Provider | Type | Relays | Context | Max output | Input | Output | Capabilities | Price (input / output) | Details | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
Hy4hy4 | Tencent | Text | 6 sites | 1.05M | 65.5K | Text: SupportedImage: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | OfficialNo priceLowest$0.0059 / $0.0175 | 2026-11-04 | |
A next-generation multimodal productivity model with significantly enhanced Agent and complex task execution capabilities. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
GPT-6.1 Terragpt-6.1-terra | OpenAI | Text | 75 sites | 1.05M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | 51.3 | Official$1 / $5Lowest$0.003 / $0.015 | 2026-10-26 | |
GPT-6.1 Terra is the mid-tier model in OpenAI's GPT-6.1 series, built for everyday computer use and agent tasks such as reading screens, clicking through apps, browsing and drafting messages. It accepts text and image input, with about a 1.05M context window, up to 128K output, and reasoning efforts from none to max. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Kimi K3.1 Prokimi-k3.1-pro | Moonshot AI | Text | 49 sites | 1.05M | 131.1K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | 46.5 | Official$2 / $10Lowest$0.0299 / $0.1496 | 2026-10-23 | |
Multimodal Kimi model with 1M context and always-on reasoning for coding, 3D creation, financial research, and agentic work Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
MiniMax-M3.1-Flashminimax-m3.1-flash | MiniMax | Text | 20 sites | 1M | 512K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | 29.1 | Official$0.15 / $0.6Lowest$0.0075 / $0.0299 | 2026-10-19 | |
MiniMax-M3.1-Flash launched on 2026-10-19 and is the current text model on M Plan and pay-as-you-go. MiniMax describes it as a native multimodal model with a 1M context window and adjustable thinking depth, for coding, agents, and computer use. Pay-as-you-go pricing is ¥1/M input, ¥4/M output, and ¥0.1/M cache read. The Preview name is retired; those requests are served by this model (verified 2026-10-20). Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
DeepSeek V4.1 Flash 1017deepseek-v4.1-flash-1017 | DeepSeek | Text | 22 sites | 1M | 384K | Text: SupportedImage: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | - | Official$0.15 / $0.6Lowest$0.0017 / $0.0069 | 2026-10-17 | |
DeepSeek-V4.1-Flash-1017 is the V4.1 Flash version released on 2026-10-17. The website, app, and API now serve this version, while the API name remains deepseek-flash. Context is 1M and max output is 384K. Requests to the retired names deepseek-v4-flash and deepseek-v4-flash-vision-exp are also served by this version at Flash pricing. The release note highlights coding, building projects from scratch, and tuning for DeepSeek Harness, CodeBuddy, and WorkBuddy (verified 2026-10-20). Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Gemini 4 Argongemini-4-argon | Text | 66 sites | 1.05M | 65.5K | Text: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | 50.4 | Official≈$2 / $10.02Lowest$0.018 / $0.0898 | 2026-10-13 | ||
Gemini 4 Argon is Google's flagship reasoning model for complex software engineering, math and science problems, and multi-step agents. It accepts text, image, video, audio, and PDF input, shares a 1,048,576-token window across input, thinking, and output, and supports long decoding with low, medium, or high thinking. Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Muse Wondermuse-wonder-1.0 | Meta | Text | 3 sites | 1.05M | 131.1K | Text: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | 43.6 | Official$4 / $20Lowest$1.47 / $7.35 | 2026-10-12 | |
Muse Wonder is Meta's frontier multimodal reasoning model for agents, long-running coding, and computer use, with memory across conversations. The Cursor model id is muse-wonder-1.0, with a 300K default context and a 1M maximum. Pricing is $4/M input, $0.20/M cache read, and $20/M output. There is no cache-write price and no long-context multiplier above 300K. Legacy Cursor plans draw it from the Other Models pool (verified 2026-10-20). Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Claude Haiku 5.5claude-haiku-5-5 | Anthropic | Text | 79 sites | 1M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | 36.5 | Official$0.1 / $0.5Lowest$0.0018 / $0.009 | 2026-10-07 | |
Claude Haiku 5.5 is for high-volume, latency-sensitive work such as classification, extraction, and routing. The API id is claude-haiku-5-5. It accepts text and image input, with a 1M context window, up to 128K output tokens, and adaptive thinking at a default effort of medium. The overview lists pricing from $0.10/M input and $0.50/M output. The Cursor price table lists $0.10 input, $0.125 cache write, $0.01 cache read, and $0.50 output (verified 2026-10-09). Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Nano Banana 2.1gemini-nano-banana-2.1 | Image | 26 sites | 131.1K | 32.8K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: SupportedImage: Supported | Reason: SupportedTools: Not supportedJSON: UnknownTemperature: Not supported | - | OfficialNo priceLowest$0.0031/ 次 | 2026-10-06 | ||
Image model for prompt-driven generation, multi-turn editing, and visual design workflows Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: Not supportedStructured output: UnknownTemperature: Not supported Pricing
/ 次 | ||||||||||||
GPT 6.1 Solgpt-6.1-sol | OpenAI | Text | 111 sites | 1.05M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | 54.6 | Official$2 / $10Lowest$0.006 / $0.0299 | 2026-09-29 | |
GPT-6.1 Sol is OpenAI's model for complex coding, computer use, and professional work, positioned as near-Astra performance at a lower cost. The API id is gpt-6.1-sol. It accepts text and image input, with a 1,050,000 context window, up to 922,000 input tokens, up to 128,000 output tokens, and an April 30, 2026 knowledge cutoff. reasoning.effort supports low, medium (default), high, xhigh, and max. none and minimal are not supported. Text pricing is $2/M input, $0.10/M cached input, $2.50/M cach Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Sonnet 5.5claude-sonnet-5-5 | Anthropic | Text | 104 sites | 1M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | 54.5 | Official$2 / $10Lowest$0.018 / $0.0898 | 2026-09-28 | |
Claude Sonnet 5.5 is the current Anthropic balance of speed and intelligence for everyday coding, agents, and knowledge work. The API id is claude-sonnet-5-5. It accepts text and image input, with a 1M context window, up to 128K output tokens, and adaptive thinking at a default effort of high. API pricing is $2/M input and $10/M output (verified 2026-10-09). Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
GPT 6 Lunagpt-6-luna | OpenAI | Text | 50 sites | 1.05M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | 27.9 | Official$0.1 / $0.5Lowest$0.0015 / $0.0075 | 2026-09-22 | |
GPT-6 Luna is OpenAI's most efficient model for focused, high-volume tasks. It accepts text and image input, with about a 1.05M context window, up to 128K output tokens, and reasoning efforts from none to max. Best for batch requests, everyday development, and high-throughput production use. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
GPT 6 Solgpt-6-sol | OpenAI | Text | 110 sites | 1.05M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | 44.2 | Official$2 / $10Lowest$0.006 / $0.0299 | 2026-09-22 | |
GPT-6 Sol is OpenAI's model for complex coding and agentic workflows. It accepts text and image input, with about a 1.05M context window, up to 128K output tokens, and reasoning efforts from none to max. Best for production coding that needs tool use, computer use, and multi-step work. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Opus 5.5claude-opus-5-5 | Anthropic | Text | 106 sites | 1M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | 59.1 | Official$4 / $20Lowest$0.0359 / $0.1795 | 2026-09-22 | |
Claude Opus 5.5 is Anthropic's model for long-running agentic coding and knowledge work, and the recommended starting point for most workloads. It accepts text and image input, with a 1M context window, up to 128K output tokens, and adaptive thinking always on. Best for long-horizon coding, knowledge work, and production tasks that need reliable tool use. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
MiMo-V2.6-Promimo-v2.6-pro | Xiaomi | Text | 19 sites | 1.05M | 131.1K | Text: SupportedImage: SupportedAudio: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | 29.9 | Official$0.435 / $0.87Lowest$0.0058 / $0.0117 | 2026-09-22 | |
MiMo-V2.6-Pro is Xiaomi's native omnimodal model released and open-sourced on 2026-09-22. The API name is mimo-v2.6-pro. The release note says API pricing stays on the V2.5 schedule. As of 2026-10-09, China realtime pricing is ¥0.025/M cache hit, ¥3/M cache miss, and ¥6/M output. It is included in the Xiaomi Token Plan. The ultra-speed name mimo-v2.6-pro-ultraspeed is priced separately and is not on the Token Plan model list. Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
MiMo-V2.6-Flashmimo-v2.6-flash | Xiaomi | Text | 20 sites | 1.05M | 131.1K | Text: SupportedImage: SupportedAudio: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | 26.7 | Official$0.14 / $0.28Lowest$0.0138 / $0.0276 | 2026-09-22 | |
MiMo-V2.6-Flash is Xiaomi's native omnimodal fast model released with Pro on 2026-09-22. The API name is mimo-v2.6-flash. The release note says it beats MiMo-V2.5-Pro on agent tasks and keeps V2.5 API pricing. As of 2026-10-09, China realtime pricing is ¥0.02/M cache hit, ¥1/M cache miss, and ¥2/M output. It is included in the Xiaomi Token Plan. Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Grok 4.7grok-4.7 | xAI | Text | 76 sites | 500K | 500K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | 35.5 | Official$2 / $6Lowest$0.006 / $0.018 | 2026-09-21 | |
Grok 4.7 is xAI's most capable model for coding and knowledge work. It uses a larger base than Grok 4.6, with longer training on multi-hour tasks, and it is better at checking its own work, managing long context, and the Grok Bot harness. Served at the same price and speed as Grok 4.6, with a 500K context window plus text and image input. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
MiMo-V2.6-Pro-UltraSpeedmimo-v2.6-pro-ultraspeed | Xiaomi | Text | 4 sites | 1.05M | 131.1K | Text: SupportedImage: SupportedAudio: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | - | Official$4.35 / $8.7Lowest$1.71 / $3.43 | 2026-09-21 | |
MiMo-V2.6-Pro-UltraSpeed is the ultra-speed mode of MiMo-V2.6-Pro. The API name is mimo-v2.6-pro-ultraspeed. Xiaomi says it reaches up to about 20x inference speed. China realtime pricing is ¥0.25/M cache hit, ¥30/M cache miss, and ¥60/M output. Batch inference is not supported. The eight models listed on the personal Token Plan do not include this name, so it is pay-as-you-go only (verified 2026-10-09). Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Qwen3.8 Omni Flashqwen3.8-omni-flash | Alibaba Cloud | Text | 5 sites | 1M | 131.1K | Text: SupportedImage: SupportedAudio: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official$0.15 / $0.47Lowest$0.0182 / $0.057 | 2026-09-17 | |
Qwen omni model for text, vision, audio, and multimodal agent tasks Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Step 5 Previewstep-5-preview | StepFun (China) | Text | 5 sites | 1M | 1M | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | 28.5 | Official$1 / $2.7Lowest$0.0077 / $0.0207 | 2026-09-16 | |
Step 5 Preview is StepFun's next flagship base model for real tasks, and it is listed on Step Plan. It accepts text, image, and video input, with a 1M token context window (verified 2026-10-20). Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Qwen-Image-2.1qwen-image-2.1 | Alibaba Cloud | Image | 2 sites | 8.2K | - | Text: SupportedImage: SupportedFile: Supported | Image: Supported | Reason: Not supportedTools: Not supportedJSON: UnknownTemperature: Not supportedOpen weights: Supported | - | OfficialNo priceLowest$0.0264/ 次 | 2026-09-14 | |
Unified text-to-image generation and image editing model with native RGBA transparency and reference-image editing Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Supported OutputText: Not supportedImage: SupportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: Not supportedTool call: Not supportedStructured output: UnknownTemperature: Not supported Pricing
/ 次 | ||||||||||||
Kimi K2.8 Previewkimi-k2.8-preview | Moonshot AI | Text | 9 sites | 1.05M | - | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | - | OfficialNo priceLowest$0.0264 / $0.1018 | 2026-09-11 | |
Kimi K2.8 Preview is a Kimi Code programming model. The API id is kimi-for-coding. It is available on Andante / Plus and above, with up to a 1M context window and thinking levels low, high, and max (verified 2026-10-20). Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
GPT Image 2.5 Flaregpt-image-2.5-flare | OpenAI | Image | 66 sites | - | - | Text: SupportedImage: SupportedFile: Supported | Image: Supported | Reason: Not supportedTools: Not supportedJSON: UnknownTemperature: Not supported | - | OfficialNo priceLowest$0.0022/ 次 | 2026-09-08 | |
Image model for prompt-driven generation, editing, and visual design workflows Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Supported OutputText: Not supportedImage: SupportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: Not supportedTool call: Not supportedStructured output: UnknownTemperature: Not supported Pricing
/ 次 | ||||||||||||
GPT Image 2.5 Sunburstgpt-image-2.5-sunburst | OpenAI | Image | 66 sites | - | - | Text: SupportedImage: SupportedFile: Supported | Image: Supported | Reason: Not supportedTools: Not supportedJSON: UnknownTemperature: Not supported | - | OfficialNo priceLowest$0.0022/ 次 | 2026-09-08 | |
Image model for prompt-driven generation, editing, and visual design workflows Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Supported OutputText: Not supportedImage: SupportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: Not supportedTool call: Not supportedStructured output: UnknownTemperature: Not supported Pricing
/ 次 | ||||||||||||
GPT-6 Astra (Fast)gpt-6-astra-fast | OpenAI | Text | 2 sites | 1.05M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | - | OfficialNo priceLowest$0.2992 / $1.5 | 2026-09-04 | |
Fast variant of GPT-6 Astra for low-latency assistance and high-volume workloads. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
GPT-6 Astragpt-6-astra | OpenAI | Text | 117 sites | 1.05M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | 56.9 | Official$10 / $50Lowest$0.0299 / $0.1496 | 2026-09-03 | |
GPT-6 Astra is OpenAI's next-generation flagship for complex reasoning, long-horizon coding, computer use, and research, with about 1M context and low–max reasoning efforts. Best for production workflows that need high-quality output, tool use, and multi-step reasoning. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Muse Spark 1.3muse-spark-1.3 | Meta | Text | 6 sites | 1.05M | 131.1K | Text: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | 36.6 | Official$1.25 / $4.25Lowest$0.0224 / $0.0449 | 2026-09-02 | |
Muse Spark 1.3 is a multimodal reasoning model released by Meta Superintelligence Labs on 2026-09-02 with 1M context and text/image/video/document input; Meta Model API offers muse-spark-1.3 and muse-spark-1.3-contributor (lower price in exchange for data contribution), and it is already live in the Muse Code coding agent. Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Qwen3.8 Max 0902qwen3.8-max-0902 | Alibaba Cloud | Text | 6 sites | 1M | 131.1K | Text: SupportedImage: SupportedVideo: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | 33.7 | OfficialNo priceLowest$0.2154 / $0.6463 | 2026-09-02 | |
2026-09-02 upgraded snapshot of Qwen3.8 Max with stronger coding, collaborative agents, and multimodal document understanding Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
DeepSeek V4.1 Flashdeepseek-v4.1-flash | DeepSeek | Text | 79 sites | 1M | 384K | Text: SupportedImage: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | 28.6 | Official$0.15 / $0.6Lowest$0.0017 / $0.0069 | 2026-09 | |
The official API name for DeepSeek V4.1 Flash remains deepseek-flash. As of 2026-10-17 that name is served by DeepSeek-V4.1-Flash-1017, with the same calling method. It has a 1M context window, up to 384K output, thinking and non-thinking modes, tool calls, and image understanding. Flash prices in CNY per million tokens, off-peak / peak, are 0.02 / 0.04 cache hit, 1 / 2 cache miss, and 4 / 8 output. Peak hours are Beijing weekdays 09:00–12:00 and 14:00–18:00, excluding mainland public holidays ( Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Gemini 3.8 Flashgemini-3.8-flash | Text | 76 sites | 1.05M | 65.5K | Text: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | 30.8 | Official$0.75 / $3.75Lowest$0.0067 / $0.0337 | 2026-09 | ||
Gemini 3.8 Flash is Google's most capable Flash-class model, with clear gains over 3.7 Flash in software engineering, agentic tasks, and multi-step reasoning, plus 1M context and multimodal input. Best for everyday development and agent workflows that need faster responses, tool use, and long context. Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Claude Mythos 5.1claude-mythos-5-1 | Anthropic | Text | 2 sites | 1M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | - | OfficialNo priceLowest$0.4488 / $2.24 | 2026-09-01 | |
Restricted Claude model for advanced cybersecurity and biology research workflows Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Gemini Omni 1.1 Flashgemini-omni-1.1-flash | Video | 2 sites | 1.05M | 57.9K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Video: Supported | Reason: SupportedTools: Not supportedJSON: UnknownTemperature: Not supported | - | OfficialNo priceLowest$0.1737/ 次 | 2026-08-27 | ||
Video generation and editing model for scene extension, frame interpolation, and reference-guided multimodal control Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: Not supportedImage: Not supportedAudio: Not supportedVideo: Supported API capabilitiesReasoning: SupportedTool call: Not supportedStructured output: UnknownTemperature: Not supported Pricing
/ 次 | ||||||||||||
Qwen3.8 Flash Nextqwen3.8-flash-next | Alibaba Cloud | Text | 1 site | 262.1K | 131.1K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | - | OfficialNo priceLowest$9.19 / $9.19 | 2026-08-27 | |
Open-weight experimental preview of the Qwen4 architecture: hybrid-attention MoE (125B total, 6B active) with vision encoder for coding, agent tasks, and image and video understanding Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
DeepSeek V4 Flash Vision Expdeepseek-v4-flash-vision-exp | DeepSeek | Text | 9 sites | 1M | 384K | Text: SupportedImage: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | - | Official$0.15 / $0.6Lowest$0.007 / $0.0209 | 2026-08-21 | |
Experimental multimodal DeepSeek V4 Flash model for image understanding, coding, and agentic work Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Qwen3.8 27Bqwen3.8-27b | Alibaba Cloud | Text | 5 sites | 262.1K | 32.8K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | 20.1 | OfficialNo priceLowest$0.012 / $0.0718 | 2026-08-14 | |
Dense 27B vision-language model for coding, agent tasks, and image and video understanding Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Gemini 3.7 Flashgemini-3.7-flash | Text | 47 sites | 1.05M | 65.5K | Text: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official≈$0.8079 / $4.04Lowest$0.0067 / $0.0337 | 2026-08-13 | ||
Gemini 3.7 Flash is a Google multimodal model released on 2026-08-13 for fast agentic workflows, coding, tool use, and complex multi-step reasoning, with a 1,048,576-token context window. Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
DeepSeek V4 Prodeepseek-v4-pro | DeepSeek | Text | 76 sites | 1M | 384K | Text: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | 23.1 | Official≈$0.4488 / $0.8977Lowest$0.0062 / $0.0186 | 2026-08-13 | |
DeepSeek V4 Pro is a high-performance model for complex reasoning and agentic coding, with 1M context. Best for stronger reasoning, rich world knowledge, competitive coding, and long-horizon agent tasks. Basics
Limits
InputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Not supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Gemini Flash Latestgemini-flash-latest | Text | 1 site | 1.05M | 65.5K | Text: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official$0.75 / $3.75Lowest$0.0202 / $0.1697 | 2026-08-13 | ||
High-efficiency Gemini model for agentic workflows, coding, and multimodal reasoning Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Grok 4.6grok-4.6 | xAI | Text | 76 sites | 500K | 500K | Text: SupportedImage: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official≈$2.15 / $6.46Lowest$0.006 / $0.018 | 2026-08-12 | |
Grok 4.6 is an xAI multimodal frontier model for long-running agents, coding, knowledge work, reasoning, tool use, and visual projects, with a 500K context window. Best for production jobs that need long-horizon execution, tool collaboration, and mixed text/visual input. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
DeepSeek V4 Pro 0813deepseek-v4-pro-0813 | DeepSeek | Text | 31 sites | 1M | 384K | Text: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | - | Official$0.66 / $1.98Lowest$0.0062 / $0.0186 | 2026-08-12 | |
DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes Basics
Limits
InputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Not supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
GPT-5.6 Cybergpt-5.6-cyber | OpenAI | Text | 1 site | 400K | 128K | Text: SupportedImage: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | - | Official$12.5 / $75Lowest$0.206 / $1.23 | 2026-08-07 | |
Cybersecurity model for authorized vulnerability research and security testing Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Grok Imagine Image 2.0grok-imagine-image-2.0 | xAI | Image | 23 sites | 64K | - | Text: SupportedImage: SupportedFile: Supported | Image: Supported | Reason: Not supportedTools: Not supportedJSON: UnknownTemperature: Not supported | - | OfficialNo priceLowest$0.0019/ 次 | 2026-08-07 | |
Image model for prompt-driven generation, editing, and visual design workflows Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Supported OutputText: Not supportedImage: SupportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: Not supportedTool call: Not supportedStructured output: UnknownTemperature: Not supported Pricing
/ 次 | ||||||||||||
Muse Spark 1.2muse-spark-1.2 | Meta | Text | 3 sites | 1.05M | 131.1K | Text: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official$1.25 / $4.25Lowest$0.0224 / $0.0449 | 2026-08-05 | |
Muse Spark 1.2 is a multimodal reasoning model released by Meta on 2026-08-05 with 1M context, launched together with the Muse Code terminal coding agent and parallel sub-agents. Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Grok Voice STT 1.0grok-voice-stt-1.0 | xAI | Text | 1 site | 15K | 15K | Audio: SupportedFile: Supported | Text: Supported | Reason: Not supportedTools: Not supportedJSON: UnknownTemperature: Not supported | - | OfficialNo priceLowest$0.4384 / - | 2026-08-04 | |
Grok Voice STT 1.0 is xAI's speech-to-text model. It supports transcription with word-level timestamps, optional speaker diarization, and multichannel audio. Basics
Limits
InputText: Not supportedImage: Not supportedAudio: SupportedVideo: Not supportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: Not supportedTool call: Not supportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Claude Fable 5.1claude-fable-5.1 | Anthropic | Text | 104 sites | - | - | - | Text: Supported | Reason: UnknownTools: UnknownJSON: UnknownTemperature: Unknown | 50.0 | OfficialNo priceLowest$0.0898 / $0.4488 | 2026-08 | |
Claude Fable 5.1 improves on Claude Fable 5 across the board, with the biggest gains in agentic coding, long-running agentic workflows and tool use; 1M context, priced at $10/M input and $50/M output. Currently ranked #1 on all three Artificial Analysis indices: Intelligence 53.4, Coding 81.6, Agentic 58.0. Basics
Limits
InputText: Not supportedImage: Not supportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Unknown OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: UnknownTool call: UnknownStructured output: UnknownTemperature: Unknown Pricing
| ||||||||||||
Qwen3.8 Maxqwen3.8-max | Alibaba Cloud | Text | 42 sites | 1M | 131.1K | Text: SupportedImage: SupportedVideo: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | 53.4 | Official$2 / $6Lowest$0.017 / $0.0509 | 2026-08 | |
Qwen3.8 Max is Alibaba Cloud Model Studio's Qwen 3.8 flagship model, available on Token Plan personal and team tiers; qwen3.8-max-preview has been retired and auto-routes here. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Qwen3.8 Flashqwen3.8-flash | Alibaba Cloud | Text | 29 sites | 1M | 131.1K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official$0.15 / $0.47Lowest$0.0055 / $0.0162 | 2026-08 | |
Qwen3.8 Flash is the fast variant in the Qwen 3.8 series, available on Token Plan personal and team tiers for high-frequency and lighter workloads. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
GLM 5.3glm-5.3 | Zhipu AI | Text | 72 sites | 1M | 131.1K | Text: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | 31.7 | Official≈$1.2 / $4.19Lowest$0.0314 / $0.0987 | 2026-08 | |
GLM 5.3 is Zhipu's next-generation flagship for reasoning, tool use, and long-horizon coding, with open weights and about 1M context. Best for engineering work, agent orchestration, and production jobs that need stable long-context execution; verify exact capabilities and endpoints with the official channel. Basics
Limits
InputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Not supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
GLM 5.3 Flashglm-5.3-flash | Zhipu AI | Text | 68 sites | 1M | 131.1K | Text: SupportedImage: SupportedVideo: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | 27.8 | Official$0.15 / $0.5Lowest$0.0034 / $0.0112 | 2026-08 | |
GLM-5.3-Flash is the first native multimodal model in Zhipu's GLM-5 series (320B total / 18B activated), with hybrid linear and sparse attention, image/video/file input, 1M context, and 128K max output. It is available on all GLM Coding Plan tiers at roughly one-third the credit cost of GLM-5.3. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Hy4 Previewhy4-preview | Tencent | Text | 18 sites | 1.02M | 64K | Text: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | - | OfficialNo priceLowest$0.009 / $0.0269 | 2026-08 | |
Hy4 preview is Tencent Hunyuan's next-generation preview model, now listed on Tencent Cloud TokenHub general and Hy Token Plans; refer to official docs for capabilities and context limits. Basics
Limits
InputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Not supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
DeepSeek V4 Flash 0731deepseek-v4-flash-0731 | DeepSeek | Text | 26 sites | 1M | 384K | Text: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | - | OfficialNo priceLowest$0.0086 / $0.0259 | 2026-07-31 | |
Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding Basics
Limits
InputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Not supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Grok Voice TTS 1.0grok-voice-tts-1.0 | xAI | Audio | 1 site | 15K | 15K | Text: Supported | Audio: Supported | Reason: Not supportedTools: Not supportedJSON: UnknownTemperature: Not supported | - | OfficialNo priceLowest$3.83 / - | 2026-07-31 | |
Convert text into spoken audio with a single API call. The API supports a rich set of expressive voices, inline speech tags for fine-grained delivery control, and output formats from high-fidelity MP3 to telephony-optimized μ-law. Basics
Limits
InputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Not supported OutputText: Not supportedImage: Not supportedAudio: SupportedVideo: Not supported API capabilitiesReasoning: Not supportedTool call: Not supportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
MiniMax H3minimax-h3 | MiniMax | Video | 11 sites | - | - | Text: SupportedImage: SupportedAudio: SupportedVideo: SupportedFile: Supported | Audio: SupportedVideo: Supported | Reason: Not supportedTools: Not supportedJSON: UnknownTemperature: Not supportedOpen weights: Supported | - | OfficialNo priceLowest$0.0025/ 次 | 2026-07-31 | |
MiniMax general-purpose video model with multimodal references and synchronized stereo audio. Open H3-Base checkpoints; Context-IR and 2K regeneration are not included in the initial weights release. Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: Not supportedImage: Not supportedAudio: SupportedVideo: Supported API capabilitiesReasoning: Not supportedTool call: Not supportedStructured output: UnknownTemperature: Not supported Pricing
/ 次 | ||||||||||||
Opus 5claude-opus-5 | Anthropic | Text | 108 sites | 1M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | - | Official$5 / $25Lowest$0.0207 / $0.1036 | 2026-07-24 | |
Claude Opus 5 is Anthropic's next-generation Opus flagship for agentic coding, hard reasoning, and long-running workflows, with adaptive thinking enabled by default. Best for highly autonomous production work that needs reliable engineering output and deep analysis. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Gemini 3.5 Flash Litegemini-3.5-flash-lite | Text | 27 sites | 1.05M | 65.5K | Text: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | 11.7 | Official$0.3 / $2.5Lowest$0.0073 / $0.0605 | 2026-07-21 | ||
Fast Gemini model balancing multimodal reasoning, tool use, and cost Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Gemini Flash-Lite Latestgemini-flash-lite-latest | Text | 3 sites | 1.05M | 65.5K | Text: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official$0.3 / $2.5Lowest$0.0081 / $0.0337 | 2026-07-21 | ||
Fast Gemini model balancing multimodal reasoning, tool use, and cost Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Qwen3.8 Max Previewqwen3.8-max-preview | Alibaba Cloud | Text | 3 sites | 1M | 131.1K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | OfficialNo priceLowest$0.2094 / $0.6282 | 2026-07-19 | |
Preview Qwen flagship for million-token multimodal reasoning and long-horizon agentic workflows Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Kimi K3kimi-k3 | Moonshot AI | Text | 74 sites | 1.05M | 131.1K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supportedOpen weights: Supported | 17.4 | Official≈$2.99 / $14.96Lowest$0.0449 / $0.2244 | 2026-07-17 | |
Kimi K3 is Moonshot AI's most capable flagship, with 2.8 trillion parameters, Kimi Delta Attention, Attention Residuals, sparse MoE, native vision, and a 1M-token context window. It targets long-horizon coding, knowledge work, and reasoning, with tool calling, structured output, and automatic context caching. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Grok 4.5grok-4.5 | xAI | Text | 77 sites | 500K | 500K | Text: SupportedImage: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official≈$2.05 / $6.15Lowest$0.006 / $0.018 | 2026-07-16 | |
Grok 4.5 is xAI's frontier model for coding, agentic tasks, and knowledge work. It has a 500K context window, low/medium/high reasoning effort, and supports function calling, web search, X search, and code execution. Long-context pricing applies when the prompt reaches 200K tokens. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Inklinginkling | Thinking Machines | Text | 2 sites | 1.05M | 1.05M | Text: SupportedImage: SupportedAudio: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | 12.5 | Official$1.87 / $4.68Lowest$1.94 / $4.85 | 2026-07-15 | |
Multimodal MoE reasoning model (975B total, 41B active) for text, image, and audio Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: Not supportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Qwen3.7 Flashqwen3.7-flash | Alibaba Cloud | Text | 4 sites | 1M | 131.1K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official$0.03 / $0.13Lowest$0.0148 / $0.0592 | 2026-07-15 | |
Lightweight multimodal Qwen model for high-throughput text, image, and video tasks Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
GPT-5.6 Lunagpt-5.6-luna | OpenAI | Text | 74 sites | 1.05M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | - | Official$0.2 / $1.2Lowest$0.0016 / $0.0096 | 2026-07-09 | |
Cost-efficient GPT-5.6 model for fast, high-volume workloads Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
GPT 5.6 Solgpt-5.6-sol | OpenAI | Text | 119 sites | 1.05M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | - | Official$4 / $20Lowest$0.015 / $0.0898 | 2026-07-09 | |
GPT 5.6 Sol is the high-performance GPT-5.6 variant for complex reasoning, long-horizon coding, and agentic workflows, with about 1M context. Best for production use that needs reliable output quality, tool use, and multi-step reasoning. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
GPT 5.6 Terragpt-5.6-terra | OpenAI | Text | 115 sites | 1.05M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | 32.3 | Official$2 / $12Lowest$0.006 / $0.0359 | 2026-07-09 | |
GPT 5.6 Terra is the lower-priced GPT-5.6 variant for everyday development, batch workloads, and cost-sensitive tasks, with about 1M context. It keeps GPT-5.6 capabilities while fitting high-frequency, lighter tasks better. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Muse Spark 1.1muse-spark-1.1 | Meta | Text | 3 sites | 1.05M | 131.1K | Text: SupportedImage: SupportedVideo: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official$1.25 / $4.25Lowest$0.6571 / $2.24 | 2026-07-09 | |
Muse Spark 1.1 is a multimodal reasoning model released by Meta on 2026-07-09 with 1M context, improved tool use and computer operation, and the Meta Model API public beta launched alongside it. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Hy3hy3 | Tencent | Text | 19 sites | 256K | 128K | Text: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | - | OfficialNo priceLowest$0.0018 / $0.0071 | 2026-07-06 | |
Tencent Hy reasoning model for coding, instruction following, and agent tasks Basics
Limits
InputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Not supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
GPT-Realtime-2.1gpt-realtime-2.1 | OpenAI | Audio | 2 sites | 128K | 32K | Text: SupportedImage: SupportedAudio: SupportedFile: Supported | Text: SupportedAudio: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | - | Official$4 / $24Lowest$0.0606 / $0.3609 | 2026-07-06 | |
Realtime speech-to-speech model with configurable reasoning, tool use, and robust voice-agent behavior Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: Not supportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: SupportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Gemini 3.6 Flashgemini-3.6-flash | Text | 46 sites | 1.05M | 65.5K | Text: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official$0.75 / $3.75Lowest$0.0135 / $0.0673 | 2026-07 | ||
Gemini 3.6 Flash is a Google multimodal model for fast responses and high-throughput workloads. It provides a 1,048,576-token context window for coding, agentic workflows, and long-context processing. Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Claude Sonnet 5claude-sonnet-5 | Anthropic | Text | 105 sites | 1M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | - | Official$2 / $10Lowest$0.0124 / $0.0621 | 2026-06-30 | |
Claude Sonnet 5 balances speed and output quality, with adaptive thinking enabled by default and lower official pricing than Opus and Fable, plus about 200K context. Best for everyday development, code review, documentation, and medium-complexity tasks. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Gemini Omni Flash Previewgemini-omni-flash-preview | Video | 4 sites | 1.05M | 57.9K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Video: Supported | Reason: SupportedTools: Not supportedJSON: UnknownTemperature: Not supported | - | Official$1.5 / $17.5Lowest$0.6732 / $4.04 | 2026-06-30 | ||
Video generation and editing model for fast, conversational text- and image-to-video workflows Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: Not supportedImage: Not supportedAudio: Not supportedVideo: Supported API capabilitiesReasoning: SupportedTool call: Not supportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
LongCat-2.0longcat-2.0 | 美团 | Text | 3 sites | 1M | 131.1K | Text: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | - | OfficialNo priceLowest$0.0673 / $0.2693 | 2026-06-30 | |
Meituan LongCat-2.0, a reasoning model with tool calling and a 1M-token context window Basics
Limits
InputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Not supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Nano Banana 2 Litegemini-3.1-flash-lite-image | Image | 6 sites | 65.5K | 4.1K | Text: SupportedImage: SupportedFile: Supported | Text: SupportedImage: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official$0.25 / $30Lowest$0.0673 / $0.6732 | 2026-06-30 | ||
Fastest, most cost-efficient Gemini image model for high-volume 1K generation and editing Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Supported OutputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Seed 2.1 Proseed-2.1-pro | ByteDance | Text | 5 sites | 256K | 256K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official$0.8906 / $4.45Lowest$0.2222 / $1.11 | 2026-06-23 | |
Flagship ByteDance Seed 2.1 model for complex multimodal reasoning, coding, and agents Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Seed 2.1 Turboseed-2.1-turbo | ByteDance | Text | 5 sites | 256K | 256K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official$0.4453 / $2.23Lowest$0.1111 / $0.5554 | 2026-06-23 | |
Faster ByteDance Seed 2.1 model for multimodal reasoning and latency-sensitive agent workflows Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Seed Characterseed-character | ByteDance | Text | 2 sites | 256K | 256K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official$0.1187 / $0.2969Lowest$11.22 / $11.22 | 2026-06-23 | |
ByteDance Seed model optimized for character-driven dialogue and consistent conversational behavior Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Seed Evolvingseed-evolving | ByteDance | Text | 2 sites | 256K | 256K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official$0.8906 / $4.45Lowest$0.7271 / $3.64 | 2026-06-23 | |
Rolling ByteDance Seed model for rapidly updated reasoning, coding, and agent capabilities Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Kimi K2.7 Codekimi-k2.7-code | Moonshot AI | Text | 32 sites | 262.1K | 262.1K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supportedOpen weights: Supported | - | Official$0.95 / $4Lowest$0.0384 / $0.1745 | 2026-06-12 | |
Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Kimi K2.7 Code Highspeedkimi-k2.7-code-highspeed | Moonshot AI | Text | 4 sites | 262.1K | 262.1K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supportedOpen weights: Supported | - | Official$1.9 / $8Lowest$0.3052 / $1.83 | 2026-06-12 | |
Kimi K2.7 Code HighSpeed is the low-latency Kimi Code variant. The API id is kimi-for-coding-highspeed. It is available on Allegretto / Pro and above, for interactive editing and coding-agent loops (verified 2026-10-20). Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Claude Fable 5claude-fable-5 | Anthropic | Text | 111 sites | 1M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | - | Official$10 / $50Lowest$0.0898 / $0.4488 | 2026-06-09 | |
Claude Fable 5 is Anthropic's flagship for agentic coding, long-running workflows, and tool use, with adaptive thinking enabled by default. Best for highly autonomous tasks, complex orchestration, and reliable engineering output, and the baseline for comparing Fable 5.1. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Claude Mythos 5claude-mythos-5 | Anthropic | Text | 3 sites | 1M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | - | OfficialNo priceLowest$0.276 / $1.38 | 2026-06-09 | |
Restricted Claude model for advanced cybersecurity and biology research workflows Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Gemini 3.5 Live Translate Previewgemini-3.5-live-translate-preview | Audio | 1 site | 131.1K | 65.5K | Audio: SupportedFile: Supported | Text: SupportedAudio: Supported | Reason: Not supportedTools: Not supportedJSON: UnknownTemperature: Not supported | - | Official$3.5 / $21Lowest$2.68 / $16.08 | 2026-06-09 | ||
Low-latency audio-to-audio model for real-time speech translation across 70+ languages Basics
Limits
InputText: Not supportedImage: Not supportedAudio: SupportedVideo: Not supportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: SupportedVideo: Not supported API capabilitiesReasoning: Not supportedTool call: Not supportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Qwen3.7 Plusqwen3.7-plus | Alibaba Cloud | Text | 16 sites | 1M | 64K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official$0.4 / $1.6Lowest$0.0156 / $0.0626 | 2026-06-02 | |
Multimodal Qwen workhorse for long-context agents, visual inputs, and coding Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
MiniMax M3minimax-m3 | MiniMax | Text | 36 sites | 1.05M | 512K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | 18.9 | Official$0.3 / $1.2Lowest$0.002 / $0.0082 | 2026-06-01 | |
MiniMax multimodal model for long-context coding, perception, and agent planning Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
GLM 5.2glm-5.2 | Zhipu AI | Text | 56 sites | 1M | 131.1K | Text: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | - | Official≈$1.2 / $1.5Lowest$0.0038 / $0.0038 | 2026-06 | |
GLM-5.2 targets project-scale engineering workflows, with a truly usable 1M context window, stable long-horizon execution, and reliable adherence to engineering conventions. Best for coding, agents, and long-running development tasks. Basics
Limits
InputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Not supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Grok Imagine Video 1.5grok-imagine-video-1.5 | xAI | Video | 13 sites | 1K | - | Text: SupportedImage: SupportedAudio: SupportedFile: Supported | Video: Supported | Reason: Not supportedTools: Not supportedJSON: UnknownTemperature: Not supported | - | OfficialNo priceLowest$0.0023/ 次 | 2026-05-30 | |
Video model for image-to-video generation, editing, and extension workflows Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: Not supportedPDF: Not supportedFile: Supported OutputText: Not supportedImage: Not supportedAudio: Not supportedVideo: Supported API capabilitiesReasoning: Not supportedTool call: Not supportedStructured output: UnknownTemperature: Not supported Pricing
/ 次 | ||||||||||||
Step 3.7 Flashstep-3.7-flash | StepFun (China) | Text | 1 site | 256K | 256K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | - | Official$0.2 / $1.15Lowest$8.38 / $8.38 | 2026-05-29 | |
Newer StepFun flash model for faster agents, coding, and multimodal prompts Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Claude Opus 4.8claude-opus-4-8 | Anthropic | Text | 113 sites | 1M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | - | Official$5 / $25Lowest$0.0207 / $0.1036 | 2026-05-28 | |
Claude Opus 4.8 is the current Anthropic Opus flagship for hard reasoning, long-document analysis, and production-grade coding, with about 200K context. Best when output quality, complex problem solving, and deep analysis matter most. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
Nano Banana 2gemini-3.1-flash-image | Image | 12 sites | 131.1K | 32.8K | Text: SupportedImage: SupportedVideo: SupportedPDF: SupportedFile: Supported | Text: SupportedImage: Supported | Reason: SupportedTools: Not supportedJSON: UnknownTemperature: Supported | - | Official$0.5 / $60Lowest$0.0121 / $0.0741 | 2026-05-28 | ||
Image model for prompt-driven generation, editing, and visual design workflows Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: SupportedFile: Supported OutputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: Not supportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Nano Banana Progemini-3-pro-image | Image | 6 sites | 65.5K | 32.8K | Text: SupportedImage: SupportedFile: Supported | Text: SupportedImage: Supported | Reason: SupportedTools: Not supportedJSON: UnknownTemperature: Supported | - | Official$2 / $120Lowest$0.1575 / $0.9439 | 2026-05-28 | ||
Nano Banana Pro for higher-fidelity image generation and design-heavy edits Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Supported OutputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: Not supportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Qwen3.7 Maxqwen3.7-max | Alibaba Cloud | Text | 27 sites | 1M | 131.1K | Text: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official≈$0.8977 / $2.69Lowest$0.0168 / $0.0505 | 2026-05-20 | |
Qwen3.7 Max is Alibaba Cloud Model Studio's flagship Qwen model. It supports thinking and non-thinking modes, up to 1M input tokens, context caching, and Batch calls, with deployments across mainland China, Singapore, Japan, Germany, and the United States. The current qwen3.7-max alias is documented as equivalent to qwen3.7-max-2026-05-20. Basics
Limits
InputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Not supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Gemini 3.1 Flash Litegemini-3.1-flash-lite | Text | 30 sites | 1.05M | 65.5K | Text: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official$0.25 / $1.5Lowest$0.0061 / $0.0363 | 2026-05-07 | ||
Low-latency Gemini model for high-volume multimodal and agent workloads Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
GPT-5.5 Instantgpt-5.5-instant | OpenAI | Text | 4 sites | 400K | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | OfficialNo priceLowest$0.1646 / $0.9874 | 2026-05-05 | |
Compact GPT model for low-latency assistance and high-volume workloads Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Gemini 3.5 Flashgemini-3.5-flash | Text | 51 sites | 1.05M | 65.5K | Text: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official≈$1.57 / $9.43Lowest$0.0135 / $0.0808 | 2026-05 | ||
Gemini 3.5 Flash combines advanced capabilities with higher speed and lower cost for real-world tasks. It is designed for agentic use cases, subagent deployment, multi-step workflows, and long-running tasks at scale, especially fast agent loops involving complex coding and iteration. Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Qwen3.6 Flashqwen3.6-flash | Alibaba Cloud | Text | 6 sites | 1M | 65.5K | Text: SupportedImage: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Supported | - | Official$0.25 / $1.5Lowest$0.0997 / $0.5983 | 2026-04-27 | |
Qwen vision-language model for visual reasoning, documents, and agent tasks Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
GPT-5.5gpt-5.5 | OpenAI | Text | 71 sites | 1.05M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | - | Official$5 / $30Lowest$0.0224 / $0.1346 | 2026-04-23 | |
GPT-5.5 is OpenAI's long-context GPT flagship for professional reasoning, writing, and long-document analysis, with a 400K context window. Best for workflows that need longer inputs, stable reasoning, and high-quality text output, and the previous-generation flagship before GPT-5.6. Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
GPT-5.5 Progpt-5.5-pro | OpenAI | Text | 7 sites | 1.05M | 128K | Text: SupportedImage: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: Not supported | - | Official$30 / $180Lowest$0.763 / $4.58 | 2026-04-23 | |
Highest-accuracy GPT-5.5 tier for slower, precision-heavy reasoning and coding Basics
Limits
InputText: SupportedImage: SupportedAudio: Not supportedVideo: Not supportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
MiMo-V2.5mimo-v2.5 | Xiaomi | Text | 13 sites | 1.05M | 131.1K | Text: SupportedImage: SupportedAudio: SupportedVideo: SupportedFile: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | - | Official$0.14 / $0.28Lowest$0.0089 / $0.0178 | 2026-04-22 | |
Open MiMo model for multimodal coding agents and long-context automation Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: Not supportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
MiMo-V2.5-Promimo-v2.5-pro | Xiaomi | Text | 14 sites | 1.05M | 131.1K | Text: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | - | Official$0.435 / $0.87Lowest$0.0269 / $0.0539 | 2026-04-22 | |
Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution Basics
Limits
InputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Not supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Gemini Embedding 2gemini-embedding-2 | Text | 5 sites | 8.2K | 3.1K | Text: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported | Text: Supported | Reason: Not supportedTools: Not supportedJSON: UnknownTemperature: Not supported | - | Official$0.2 / $0Lowest$0.0081 / - | 2026-04-22 | ||
Multimodal embedding model mapping text, images, video, audio, and PDFs into a unified embedding space Basics
Limits
InputText: SupportedImage: SupportedAudio: SupportedVideo: SupportedPDF: SupportedFile: Supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: Not supportedTool call: Not supportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
MiMo-V2.5-TTSmimo-v2.5-tts | Xiaomi | Audio | 2 sites | 8.2K | 8.2K | Text: Supported | Audio: Supported | Reason: Not supportedTools: Not supportedJSON: UnknownTemperature: Not supported | - | OfficialNo priceLowest$0.031 / $0.1522 | 2026-04-22 | |
Speech generation model for controllable voice, narration, and audio delivery Basics
Limits
InputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Not supported OutputText: Not supportedImage: Not supportedAudio: SupportedVideo: Not supported API capabilitiesReasoning: Not supportedTool call: Not supportedStructured output: UnknownTemperature: Not supported Pricing
| ||||||||||||
FAQ
Which models are in the AI model library?
Public text, image, video and audio models from Anthropic, OpenAI, Google, DeepSeek, Alibaba, ByteDance and other providers. Filter by provider, model type, input and output modalities, reasoning and tool calling.
How do official prices differ from AI gateway prices?
Official prices are the provider's own API rates. Gateway prices are public quotes from third-party AI gateways; they are often cheaper but carry stability and model-substitution risks. The lowest gateway price only counts sites with a current public quote.
How is the list sorted by default?
Models with gateway prices come first, then newest releases first. Use the column headers to sort by release date or intelligence score instead.
How can I check whether a gateway serves the real model?
Use Hvoy AI's API check: enter the gateway's endpoint and key to verify that the responding model matches what is claimed, or compare uptime and model consistency on each model's gateway ranking.
Capabilities and limits come from public sources. “Unknown” means no reliable data yet, not “not supported”. Relay counts include only sites with a current public price, de-duplicated per site.
Data source: models.dev · OpenRouter
Intelligence scores come from the OpenRouter benchmark rankings (Intelligence Index); higher is smarter.