# Tokenhot

## Docs
- en [Terms of Service](https://docs.tokenhot.ai/english/2048220m0.md): 
- en [privacy agreement](https://docs.tokenhot.ai/english/2048221m0.md): 

## API Docs
- en > Image Models > DOUBAO [doubao-seedream-4-0-250828](https://docs.tokenhot.ai/english/31515275e0.md): Generic Agent Model Seed1.8.It supports text and picture input, and with its powerful multi-modal processing capabilities, it shows excellent performance in complex application scenarios such as information retrieval, code writing, and graphical user interface (GUI) interaction.Seed1.8 demonstrates outstanding performance and broad applicability across multiple Agent benchmarks.It features industry-leading GUI Agent capabilities that build on Seed1.5-VL.Seed1.8 is also prominent in search tasks, maintaining the industry's first echelon level in multiple public Agent search benchmarks, such as a score of up to 67.6 on the BrowseComp-en benchmark, surpassing other top models such as Gemini-3-Pro.In intelligent programming-related benchmarks, Seed1.8 demonstrates stability for real-world software engineering scenarios.Among the benchmarks related to economic value areas, Seed1.8 demonstrates competence in real-world high-value workflows.**Actual API Endpoint:** `POST https://tokenhot.ai/v1/images/generations/`
- en > Image Models > DOUBAO [doubao-seedream-4-5-251128](https://docs.tokenhot.ai/english/31515276e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/v1/images/generations/`
- en > Image Models > GPT [GPT Image-1.5](https://docs.tokenhot.ai/english/31515277e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/v1/images/generations/`
- en > Image Models > Gemini [gemini-2.5-flash-image](https://docs.tokenhot.ai/english/31515278e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/v1beta/models/{model}:generateContent/`
- en > Image Models > Gemini [gemini-3-pro-image-preview](https://docs.tokenhot.ai/english/31515279e0.md): The Gemini 3 Pro Image Preview revolutionizes the landscape of AI image generation.Known internally as "Nano Banana Pro", this model is the first to deeply merge the reasoning power of a large language model with image generation, enabling a previously unimaginable combination of features: native 4K resolution output, near-perfect multilingual text rendering, real-time data anchoring based on Google searches, and true multi-round conversational image editing.For development teams and businesses that need to generate hundreds or even thousands of images per day, choosing the right image generation model directly impacts product quality, development efficiency, and operational costs.
- en > Image Models > Gemini [gemini-3.1-flash-image-preview](https://docs.tokenhot.ai/english/31515280e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/v1beta/models/{model}:generateContent/`
- en > Image Models > dall [dall-e-2](https://docs.tokenhot.ai/english/31515281e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/v1/images/generations/`
- en > Image Models > dall [dall-e-3](https://docs.tokenhot.ai/english/31515282e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/v1/images/generations/`
- en > Video Models > DOUBAO [doubao-seedance-1-5-pro-251215 (Generate Video Task)](https://docs.tokenhot.ai/english/31515283e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/jimeng/`
- en > Video Models > DOUBAO [doubao-seedance-1-5-pro-251215 (Get video task status and results)](https://docs.tokenhot.ai/english/31515284e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/v1/video/generations/{task_id}`
- en > Video Models > DOUBAO [doubao-seedance-2-0-260128](https://docs.tokenhot.ai/english/31515285e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/jimeng/`
- en > Video Models > DOUBAO [doubao-seedance-2.0](https://docs.tokenhot.ai/english/31515286e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/v1/video/generations`
- en > Video Models > DOUBAO [doubao-seedance-2.0 (Get video task information)](https://docs.tokenhot.ai/english/31515287e0.md): **Actual API Endpoint:** `GET https://tokenhot.ai/v1/video/generations/{task_id}`
- en > Video Models > DOUBAO [doubao-seedance-2.0-V2V](https://docs.tokenhot.ai/english/31515288e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/v1/video/generations`
- en > Video Models > DOUBAO [doubao-seedance-2.0-V2V (Get video task information)](https://docs.tokenhot.ai/english/31515289e0.md): * * Actual API Endpoint: * * `get https://tokenhot.ai/v1/video/generations/ {task_id}`
- en > Video Models > DOUBAO [doubao-seedance-2.0-fast](https://docs.tokenhot.ai/english/31515290e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/v1/video/generations`
- en > Video Models > DOUBAO [doubao-seedance-2.0-fast (get video task information)](https://docs.tokenhot.ai/english/31515291e0.md): **Actual API Endpoint:** `GET https://tokenhot.ai/v1/video/generations/{task_id}`
- en > Video Models > DOUBAO [doubao-seedance-2.0-fast-V2V](https://docs.tokenhot.ai/english/31515292e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/v1/video/generations`
- en > Video Models > DOUBAO [doubao-seedance-2.0-fast-V2V (Get video task information)](https://docs.tokenhot.ai/english/31515293e0.md): **Actual API Endpoint:** `GET https://tokenhot.ai/v1/video/generations/{task_id}`
- en > Video Models > Sora [Sora-2](https://docs.tokenhot.ai/english/31515294e0.md): Sora 2 is a second-generation Vincent video model launched by OpenAI. Compared with the first generation, it greatly improves the authenticity, coherence and controllability of physical simulation. It supports longer duration, movie-level multi-camera narrative and original sound and painting synchronization. It can generate high-definition video that is closer to the real physical rules and consistent across the lens of the character scene. It also supports the creation of digital doppelgänger based on short video, and the overall video creation ability is upgraded from generating simple images to closer to real-world simulation.
- en > Video Models > Sora [Sora-2 (Get Video Task Status)](https://docs.tokenhot.ai/english/31515295e0.md): **Actual API Endpoint:** `GET https://tokenhot.ai/v1/videos/{task_id}`
- en > Video Models > Sora [Sora-2 (Get Video Content)](https://docs.tokenhot.ai/english/31515296e0.md): **Actual API Endpoint:** `GET https://tokenhot.ai/v1/videos/{task_id}/content`
- en > Chat Models > Claude [claude-opus-4-6](https://docs.tokenhot.ai/english/31515297e0.md): Claude Opus 4.6 is Anthropic's most powerful model upgrade to date and a key leap in the Opus family's long-context and expert-level reasoning capabilities.It not only systematically improves core capabilities such as coding, inference, and proxy tasks, but also introduces a context window (beta) of up to 1 million tokens for the first time, significantly extending the boundaries available for models in real complex work scenarios.
- en > Chat Models > Claude [claude-opus-4-6-cc](https://docs.tokenhot.ai/english/31515298e0.md): Great for claude-code
- en > Chat Models > Claude [claude-sonnet-4-5-20250929](https://docs.tokenhot.ai/english/31515299e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/v1/messages`
- en > Chat Models > GPT [GPT 5](https://docs.tokenhot.ai/english/31515300e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/v1/chat/completions`
- en > Chat Models > GPT [GPT 5(Responses)](https://docs.tokenhot.ai/english/31515301e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/v1/responses`
- en > Chat Models > GPT [GPT 5.2](https://docs.tokenhot.ai/english/31515302e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/v1/chat/completions`
- en > Chat Models > GPT [GPT 5.2(Responses)](https://docs.tokenhot.ai/english/31515303e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/v1/responses`
- en > Chat Models > GPT [GPT 5.4](https://docs.tokenhot.ai/english/31515304e0.md): GPT5.4 is a next-generation multimodal model with superior inference capabilities, support for text and image input, networked search enhancements, and support for adjusting inference intensity.
- en > Chat Models > GPT [GPT 5.4(Responses)](https://docs.tokenhot.ai/english/31515305e0.md): GPT5.4 is a next-generation multimodal model with superior inference capabilities, support for text and image input, networked search enhancements, and support for adjusting inference intensity.
- en > Chat Models > Gemini [gemini-2.5-flash](https://docs.tokenhot.ai/english/31515306e0.md): gemini-2.5-flash
- en > Chat Models > Gemini [gemini-2.5-pro](https://docs.tokenhot.ai/english/31515307e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/v1beta/models/{model}:generateContent`
- en > Chat Models > Gemini [gemini-3-flash-preview](https://docs.tokenhot.ai/english/31515308e0.md): **Actual API Endpoint:** `POST https://tokenhot.ai/v1beta/models/{model}:generateContent`
- en > Chat Models > Gemini [gemini-3.1-pro](https://docs.tokenhot.ai/english/31515309e0.md): The core breakthrough of Gemini 3.1 Pro is its shift from static compute allocation to a dynamic “Test-Time Compute” architecture.This means that the model can intelligently allocate different computing resources and time according to the complexity of the task, rather than consuming a fixed amount of computing power for each token.This transformation is achieved through its signature three-level thinking system (Low/Medium/High).
- en > Chat Models > Gemini [gemini-3.1-pro-preview](https://docs.tokenhot.ai/english/31515310e0.md): The core breakthrough of Gemini 3.1 Pro is its shift from static compute allocation to a dynamic “Test-Time Compute” architecture.This means that the model can intelligently allocate different computing resources and time according to the complexity of the task, rather than consuming a fixed amount of computing power for each token.This transformation is achieved through its signature three-level thinking system (Low/Medium/High).