> ## Documentation Index > Fetch the complete documentation index at: https://docs.powertokens.ai/llms.txt > Use this file to discover all available pages before exploring further. # mimo-v2.5 Chat Completions > Invoke MIMO chat completions capability ## OpenAPI ````yaml api-reference/en/zmodelChat/mimo/mimo-v2.5.json POST /v1/chat/completions openapi: 3.0.1 info: title: mimo-v2.5 version: 1.0.0 description: MIMO Chat Completions documentation license: name: Project License url: https://github.com/QuantumNous/new-api/blob/main/LICENSE servers: - url: https://api.powertokens.ai description: Baze API server endpoint security: [] tags: - name: MIMO Chat description: MIMO text chat paths: /v1/chat/completions: post: tags: - MIMO Chat summary: mimo-v2.5 description: Invoke MIMO chat completions capability operationId: mimoChatCompletions_mimo_v2_5_en requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/ChatCompletionRequest' examples: Basic Call: summary: Basic text chat value: model: mimo-v2.5 messages: - role: system content: >- You are MiMo, an AI assistant developed by Xiaomi. Today is date: Tuesday, December 16, 2025. Your knowledge cutoff date is December 2024. - role: user content: please introduce yourself max_completion_tokens: 1024 temperature: 1 top_p: 0.95 stream: false stop: null frequency_penalty: 0 presence_penalty: 0 thinking: type: disabled Streaming Response: summary: Streaming chat response value: model: mimo-v2.5 messages: - role: system content: >- You are MiMo, an AI assistant developed by Xiaomi. Today is date: Tuesday, December 16, 2025. Your knowledge cutoff date is December 2024. - role: user content: - type: text text: please introduce yourself max_completion_tokens: 1024 stream: true Function Call: summary: Chat with function calling value: model: mimo-v2.5 messages: - role: system content: >- You are MiMo, an AI assistant developed by Xiaomi. Today is date: Tuesday, December 16, 2025. Your knowledge cutoff date is December 2024. - role: user content: What is the weather like in Boston today? max_completion_tokens: 1024 tools: - type: function function: name: get_current_weather description: Get the current weather in a given location parameters: type: object properties: location: type: string description: The city and state, e.g. San Francisco, CA unit: type: string enum: - celsius - fahrenheit required: - location tool_choice: auto Web Search: summary: Chat with web search value: model: mimo-v2.5 messages: - role: system content: >- You are MiMo, an AI assistant developed by Xiaomi. Today is date: Tuesday, December 16, 2025. Your knowledge cutoff date is December 2024. - role: user content: please introduce Jun Lei tools: - type: web_search max_keyword: 3 force_search: true limit: 1 user_location: type: approximate country: China region: Hubei city: Wuhan max_completion_tokens: 1024 temperature: 1 top_p: 0.95 stream: false stop: null frequency_penalty: 0 presence_penalty: 0 thinking: type: disabled Image Input: summary: Chat with image input value: model: mimo-v2.5 messages: - role: system content: >- You are MiMo, an AI assistant developed by Xiaomi. Today is date: Tuesday, December 16, 2025. Your knowledge cutoff date is December 2024. - role: user content: - type: image_url image_url: url: >- https://example-files.cnbj1.mi-fds.com/example-files/image/image_example.png - type: text text: please describe the content of the image max_completion_tokens: 1024 Audio Input: summary: Chat with audio input value: model: mimo-v2.5 messages: - role: system content: >- You are MiMo, an AI assistant developed by Xiaomi. Today is date: Tuesday, December 16, 2025. Your knowledge cutoff date is December 2024. - role: user content: - type: input_audio input_audio: data: >- https://example-files.cnbj1.mi-fds.com/example-files/audio/audio_example.wav - type: text text: please describe the content of the audio max_completion_tokens: 1024 Video Input: summary: Chat with video input value: model: mimo-v2.5 messages: - role: system content: >- You are MiMo, an AI assistant developed by Xiaomi. Today is date: Tuesday, December 16, 2025. Your knowledge cutoff date is December 2024. - role: user content: - type: video_url video_url: url: >- https://example-files.cnbj1.mi-fds.com/example-files/video/video_example.mp4 fps: 2 media_resolution: default - type: text text: please describe the content of the video max_completion_tokens: 1024 Structured Output: summary: Chat with structured output value: model: mimo-v2.5-pro messages: - role: system content: >- You are MiMo, an AI assistant developed by Xiaomi. Today is date: Tuesday, December 16, 2025. Your knowledge cutoff date is December 2024. - role: user content: >- My name is Zhang San, I am 28 years old, my email address is zhangsan@test.com, and my birthday is May 12, 1996. Please return the results in the following JSON format. Note: The birthday must be in YYYY-MM-DD format. {"name": "string or null", "age": "int or null", "email": "string or null", "birthday": "string or null"} max_completion_tokens: 1024 response_format: type: json_object Deep Thinking: summary: Chat with deep thinking value: model: mimo-v2.5-pro messages: - role: system content: >- You are MiMo, an AI assistant developed by Xiaomi. Today is date: Tuesday, December 16, 2025. Your knowledge cutoff date is December 2024. - role: user content: Introduce machine learning in three sentences. max_completion_tokens: 1024 thinking: type: enabled responses: '200': description: >- Success. Returns JSON in non-streaming mode; returns an SSE event stream in streaming mode. content: application/json: schema: $ref: '#/components/schemas/ChatCompletionResponse' examples: Basic Call: summary: Non-streaming success response value: id: 8b51f9e0515949cb8207fbd35ea6ea5c choices: - finish_reason: stop index: 0 message: content: >- Hello! I'm MiMo, Xiaomi's AI assistant created by the Xiaomi LLM-Core team. I'm here to chat, help answer questions, and assist with various tasks—whether it's providing information, brainstorming ideas, or just having a friendly conversation. Feel free to ask me anything, and I'll do my best to help! 😊 role: assistant tool_calls: null created: 1776848906 model: mimo-v2.5-pro object: chat.completion usage: completion_tokens: 72 prompt_tokens: 57 total_tokens: 129 completion_tokens_details: reasoning_tokens: 0 prompt_tokens_details: null Function Call: summary: Function calling response value: id: 8778784df68043e39a2df03258ddeaa5 choices: - finish_reason: tool_calls index: 0 message: content: '' role: assistant tool_calls: - id: call_b837e211b7764dab9b56f4d8 function: arguments: '{"location": "Boston, MA", "unit": "celsius"}' name: get_current_weather type: function reasoning_content: >- The user is asking about the current weather in Boston. I should use the get_current_weather function to get this information. Boston could refer to Boston, MA. I'll call the function with location "Boston, MA" and use celsius as the unit since that's more commonly used in many places, but actually the user didn't specify a unit. Looking at the function parameters, I can choose celsius or fahrenheit. Since Boston is in the US, maybe fahrenheit would be more appropriate? Actually, I should just pick a default or ask. But the instructions say not to ask about optional parameters if there are defaults. There's no default specified in the schema, so I should choose one. I'll go with celsius as it's more standard internationally. Let me call the function. created: 1776849125 model: mimo-v2.5-pro object: chat.completion usage: completion_tokens: 196 prompt_tokens: 349 total_tokens: 545 completion_tokens_details: reasoning_tokens: 161 prompt_tokens_details: null Web Search: summary: Web search response value: id: 828d5db56ff748d8b4623d1972c8eca2 choices: - finish_reason: stop index: 0 message: content: >- Based on the search results, here is an introduction to Jun Lei (Lei Jun): Jun Lei, also known as Lei Jun, is a renowned Chinese entrepreneur and technology industry leader. He has accumulated 30 years of entrepreneurial experience in the science and technology sector, making him a representative figure of China's internet and a well-known entrepreneur. Lei Jun was born in 1969 in Xiantao, Hubei Province. He began his entrepreneurial journey early, participating in the founding of Kingsoft (Jinshan Software) in 1992 and later serving as its CEO in 1998. That same year, he also founded Excellence Network (Zhuoyue Wang). After Kingsoft went public in 2007, he stepped down as President and CEO to take on the role of Vice Chairman. In the following years, Lei Jun actively engaged as an angel investor, backing numerous innovative enterprises such as Fanke Chengpin (Vancl), Duowan, and Youshi Technology (UCWeb). A pivotal moment came on April 6, 2010, when he decided to embark on a new venture, establishing Xiaomi Company. The first Xiaomi smartphone was officially released on August 16, 2011. Furthermore, Lei Jun returned to Kingsoft in July 2011 to oversee its online games and antivirus software divisions, and he officially assumed the position of Chairman of Kingsoft on July 11, 2011. His career also includes involvement in investment cases for companies like Shuiyang Shares, Edianyun, Yaoshibang, Cheche, and others. role: assistant annotations: - type: url_citation url: https://en.shunwei.com/team title: Team – Shunwei summary: >- Jun LEI Founding Partner & Chairman Mr. Lei Jun, with 30 years of entrepreneurial experience in the science and technology industry, is a representative of China's Internet and a well-known entrepreneShe has been involved in investment cases such as Shuiyang Shares (SZ:300740), Edianyun (HK:02416), Yaoshibang (HK:09885), Cheche, Shansong, Royal Xiaohu, Yangteng Innovation, Zhiketong, Yinghe, Chaya site_name: en.shunwei.com publish_time: '2024-10-26T07:37:04+08:00' logo_url: >- https://th.bochaai.com/favicon?domain_url=https://en.shunwei.com/team - type: url_citation url: >- https://max.book118.com/html/2023/0519/7145010064005110.shtm title: 小米公司介绍-20230519110336.pptx-原创力文档 summary: >- 小米公司介绍第1页/共25页小米创始人:雷军Millet founder: Lei Jun1969年出生于湖北仙桃,于1992年参与创办金山软件,1998年出任金山软件CEO,1998年创办了卓越网。1969 born in Hubei Xiantao, in 1992, and participated in the founding of Kingsoft, 1998 served as CEO of golden hill software. In 1998 founded the network of excellence.2007年,金山软件上市后,雷军卸任金山软件总裁兼CEO职务,担任副董事长。2007, Kingsoft listing, Kingsoft chairman and CEO Lei Juns office, served as vice chairman.之后几年,雷军作为天使投资人,投资了凡客诚品、多玩、优视科技等多家创新型企业。After a few years, Lei Jun as an angel investor, invested more, Eslite, superior technology and a number of innovative enterprises.2010年4月6日,雷军选择重新创业,建立了小米公司,并于2011年8月16日正式发布小米手机。In April 6, 2010, Lei choose to venture, a millet company, and officially released millet mobile phone in August 16, 2011.2011年7月,雷军重返金山执掌网游与毒霸;2011年7月11日,正式担任金山软件董事长。July 2011, Lei Ju site_name: 原创力文档 publish_time: '2023-05-23T19:49:37+08:00' logo_url: >- https://th.bochaai.com/favicon?domain_url=https://max.book118.com/html/2023/0519/7145010064005110.shtm tool_calls: null created: 1776849240 model: mimo-v2.5-pro object: chat.completion usage: completion_tokens: 387 prompt_tokens: 1222 total_tokens: 1609 completion_tokens_details: reasoning_tokens: 0 prompt_tokens_details: null web_search_usage: tool_usage: 3 page_usage: 2 Image Input: summary: Image input response value: id: e27ffa1d3e31434daa71d6b6a83f6e38 choices: - finish_reason: length index: 0 message: content: >- Based on the image provided, here is a detailed description of the content: **Overall Atmosphere** The image depicts a serene, magical, and lush forest scene. It has an ethereal quality, likely enhanced by digital art techniques or AI generation, characterized role: assistant tool_calls: null reasoning_content: >- Here's a thinking process that leads to the description: 1. **Analyze the Request:** * **Input:** An image of a forest scene with a stream, sunlight, and trees. * **Task:** Describe the content of the image. 2. **Initial Impression & Overall Mood:** * The image depicts a serene, magical, and lush forest. * The lighting is dramatic, with sunbeams filtering through the canopy. * There's a sense of tranquility and nature in its undisturbed state. * The style looks somewhat photorealistic but possibly enhanced or AI-generated due to the perfect lighting and composition. 3. **Break Down Visual Elements:** * **Foreground:** * *Left side:* Large, detailed ferns. They are vibrant green and look wet or dewy. * *Bottom center/right:* A stream or small river flowing towards the viewer. * *Water:* The water is clear but reflects the light, creating a shimmering effect. It flows over smooth, moss-covered rocks. * *Rocks:* Several large, rounded stones are visible in the immediate foreground, covered in thick green moss. * **Mid-ground:** * *Trees:* Massive, ancient-looking trees with thick, gnarled trunks and exposed roots. The bark is textured and dark. * *Vegetation:* More ferns, bushes, and mossy ground cover lining the banks of the stream. * *Stream path:* The stream winds back into the forest, guiding the eye. * **Background:** * *Forest density:* The trees become denser and fade into a misty or foggy atmosphere. * *Lighting:* This is the most striking feature. "God rays" or crepuscular rays are streaming down from the upper center/right, cutting through the mist and illuminating the scene. * *Particles:* There are tiny specks floating in the light beams, looking like dust motes, pollen, or perhaps magical sparkles/fireflies. 4. **Structure the Description:** * **Introduction:** General summary of the scene (misty forest, stream, sunlight). * **The Lighting (Key Feature):** Describe the sunbeams and atmosphere. * **The Vegetation (Flora):** Mention the ferns, moss, and ancient trees. * **The Water (Hydrography):** Describe the stream, rocks, and reflections. * **Composition/Atmosphere:** Conclude with the overall feeling (peaceful, ethereal). 5. **Drafting the Description (Iterative refinement):** * *Draft 1 (Mental):* It's a forest with a river. Big trees on the sides. Sun is shining down in beams. There are ferns in the front left. The rocks are mossy. It looks magical. * *Draft 2 (More detail):* A lush forest scene. A stream flows through the center over mossy rocks. On the left, there are large ferns. The trees are old with twisted branches. The sun is breaking through the canopy, creating bright rays of light. There is mist in the background. It looks like a fantasy setting. * *Draft 3 (Polished & Descriptive - closer to final output):* * Start with the atmosphere: "This image depicts a serene and enchanting forest scene..." * Focus on the light: "...dominated by dramatic shafts of sunlight (god rays) filtering through the dense canopy..." * Describe the water: "A gentle stream flows through the center, winding over smooth, moss-covered stones..." * Describe the plants: "Lush vegetation is everywhere, particularly large, detailed ferns in the foreground on the left..." * Describe the trees: "Ancient trees with thick, gnarled trunks and exposed roots frame the scene..." 6. **Final Polish (Reviewing against the image):** * *Check:* Did I mention the moss? Yes. * *Check:* Did I mention the floating particles? Yes, add that detail about the "sparkles" or dust motes. * *Check:* Is the perspective correct? Yes, low angle looking up the stream. 7. **Final Output Generation:** (This matches the provided good response.) * "The image displays a breathtaking, ethereal forest scene..." * Break it down by lighting, vegetation, and water. * Mention the color palette (greens, browns, golden light). 8. **Formatting:** Use bullet points or paragraphs for readability. (I will use a mix of general description and specific details). created: 1776849330 model: mimo-v2.5 object: chat.completion usage: completion_tokens: 1024 prompt_tokens: 1085 total_tokens: 2109 completion_tokens_details: reasoning_tokens: 971 prompt_tokens_details: cached_tokens: 5 image_tokens: 1024 Audio Input: summary: Audio input response value: id: 87ee7097c0f7491ea27429efcc21cc26 choices: - finish_reason: stop index: 0 message: content: '' role: assistant tool_calls: null reasoning_content: >- Good morning. Could you tell me what the weather will be like today? created: 1776849418 model: mimo-v2.5 object: chat.completion usage: completion_tokens: 17 prompt_tokens: 86 total_tokens: 103 completion_tokens_details: reasoning_tokens: 15 prompt_tokens_details: audio_tokens: 25 cached_tokens: 45 Video Input: summary: Video input response value: id: d2c3bd5a08114f78a525e09b69308dc9 choices: - finish_reason: stop index: 0 message: content: >- The video features a white kitten with bright blue eyes sitting on a polished wooden floor. It is positioned near a window covered by beige curtains, with sunlight streaming in and casting shadows across the room. In the background, a green potted plant, resembling a Monstera, is visible. The kitten looks around curiously and then opens its mouth to meow. role: assistant tool_calls: null reasoning_content: >- The user wants a description of the video content. 1. **Analyze the video**: * **Subject**: A white kitten with blue eyes. * **Setting**: Indoors, near a window. There are beige curtains, a wooden floor, and a green potted plant (Monstera) in the background. * **Action**: The kitten is sitting on the wooden floor. It's looking slightly to the side. It meows. The lighting is warm and sunny, creating shadows on the floor. * **Timestamps**: * 00:00 - 00:03: The kitten sits still, looking around, then opens its mouth to meow. The sun casts shadows. 2. **Synthesize the description**: * Start with the main subject: A white kitten with blue eyes. * Describe the environment: Sitting on a wooden floor near a window with beige curtains. Sunlight is streaming in. A plant is visible in the background. * Describe the action: The kitten looks around and meows. * Mention the atmosphere: Warm, peaceful, sunny. 3. **Draft the response**: The video features a close-up of a white kitten with striking blue eyes sitting on a polished wooden floor. It is positioned near a window draped with light beige curtains. Sunlight streams through the window, casting long shadows on the floor and illuminating the kitten's fur. In the background, a green potted plant, resembling a Monstera, adds a touch of nature to the indoor setting. The kitten looks around curiously and then lets out a meow. The scene is bright and peaceful. 4. **Refine based on specific details**: * The kitten is white. * Blue eyes. * Wooden floor. * Window with curtains. * Plant in background. * Sunlight/Shadows. * Action: Meowing. 5. **Final Polish**: Combine these points into a coherent paragraph. The video shows a white kitten with blue eyes sitting on a wooden floor in a sunlit room. It's near a window with curtains and a plant in the background. The kitten meows. Let's stick to a descriptive summary. * **Subject**: White kitten, blue eyes. * **Location**: Wooden floor, near window, curtains, plant. * **Action**: Sitting, looking, meowing. * **Lighting**: Sunny, warm. Construct the final output. created: 1776849614 model: mimo-v2.5 object: chat.completion usage: completion_tokens: 590 prompt_tokens: 1260 total_tokens: 1850 completion_tokens_details: reasoning_tokens: 514 prompt_tokens_details: audio_tokens: 19 cached_tokens: 45 video_tokens: 1144 Structured Output: summary: Structured output response value: id: f88d3e6c6b0043b09871487b5c52e4b4 choices: - finish_reason: stop index: 0 message: content: >- {"name": "Zhang San", "age": 28, "email": "zhangsan@test.com", "birthday": "1996-05-12"} role: assistant tool_calls: null created: 1776850034 model: mimo-v2.5-pro object: chat.completion usage: completion_tokens: 41 prompt_tokens: 145 total_tokens: 186 completion_tokens_details: reasoning_tokens: 0 prompt_tokens_details: null Deep Thinking: summary: Deep thinking response value: id: 7f2cc80f34b541c49f5fcd40ad34bc73 choices: - finish_reason: stop index: 0 message: content: >- Machine learning is a field of artificial intelligence that enables computers to learn from data and improve their performance without being explicitly programmed for every task. It uses algorithms to identify patterns and make decisions or predictions based on input information. This approach powers everything from personalized recommendations to advanced medical diagnostics. role: assistant tool_calls: null reasoning_content: >- Hmm, the user is asking for a concise introduction to machine learning in exactly three sentences. This seems straightforward but requires careful wording to cover the core concepts without oversimplifying or overcomplicating. \n\nI should start with the basic definition, then mention how it learns from data, and end with its purpose and real-world relevance. The response should flow naturally from \"what it is\" to \"how it works\" to \"why it matters.\" \n\nKeeping it neutral and factual works best here—no need for examples or jargon. Just three clear, standalone sentences that collectively give a solid overview. created: 1776850089 model: mimo-v2.5-pro object: chat.completion usage: completion_tokens: 179 prompt_tokens: 60 total_tokens: 239 completion_tokens_details: reasoning_tokens: 121 prompt_tokens_details: null text/event-stream: schema: $ref: '#/components/schemas/ChatCompletionChunk' examples: Streaming Response: summary: Streaming response with SSE events value: >- data: {"id":"7ccecae1cc6445f395e0d41c85115663","object":"chat.completion.chunk","created":1776849042,"model":"mimo-v2.5-pro","choices":[{"index":0,"delta":{"role":"assistant","content":"","reasoning_content":null,"tool_calls":null},"logprobs":null,"finish_reason":null,"matched_stop":null}],"usage":null} data: {"id":"7ccecae1cc6445f395e0d41c85115663","choices":[{"delta":{"content":null,"role":null,"tool_calls":null,"reasoning_content":"Okay, the user"},"finish_reason":null,"index":0}],"created":1776849042,"model":"mimo-v2.5-pro","object":"chat.completion.chunk","usage":null} data: {"id":"7ccecae1cc6445f395e0d41c85115663","choices":[{"delta":{"content":null,"role":null,"tool_calls":null,"reasoning_content":" is asking"},"finish_reason":null,"index":0}],"created":1776849042,"model":"mimo-v2.5-pro","object":"chat.completion.chunk","usage":null} data: {"id":"7ccecae1cc6445f395e0d41c85115663","choices":[{"delta":{"content":null,"role":null,"tool_calls":null,"reasoning_content":" me"},"finish_reason":null,"index":0}],"created":1776849042,"model":"mimo-v2.5-pro","object":"chat.completion.chunk","usage":null} ... data: {"id":"7ccecae1cc6445f395e0d41c85115663","choices":[{"delta":{"content":"'m ready","role":null,"tool_calls":null,"reasoning_content":null},"finish_reason":null,"index":0}],"created":1776849044,"model":"mimo-v2.5-pro","object":"chat.completion.chunk","usage":null} data: {"id":"7ccecae1cc6445f395e0d41c85115663","choices":[{"delta":{"content":" to help!","role":null,"tool_calls":null,"reasoning_content":null},"finish_reason":null,"index":0}],"created":1776849044,"model":"mimo-v2.5-pro","object":"chat.completion.chunk","usage":null} data: {"id":"7ccecae1cc6445f395e0d41c85115663","object":"chat.completion.chunk","created":1776849044,"model":"mimo-v2.5-pro","choices":[{"index":0,"delta":{"role":null,"content":null,"reasoning_content":null,"tool_calls":null},"logprobs":null,"finish_reason":"stop","matched_stop":151645}],"usage":null} data: {"id":"7ccecae1cc6445f395e0d41c85115663","choices":[],"created":1776849044,"model":"mimo-v2.5-pro","object":"chat.completion.chunk","usage":{"completion_tokens":123,"prompt_tokens":55,"total_tokens":178,"completion_tokens_details":{"reasoning_tokens":71},"prompt_tokens_details":null}} data: [DONE] '400': description: Invalid request parameters. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' examples: invalid_request: summary: Parameter error value: error: message: messages is required type: invalid_request_error param: messages code: bad_request_body '401': description: Authentication failed, e.g. missing or invalid token. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' examples: unauthorized: summary: Unauthorized value: error: message: Invalid token type: api_error code: access_denied param: '' '429': description: Rate limit triggered or insufficient account quota. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' examples: quota_exceeded: summary: Insufficient quota or rate limited value: error: message: Insufficient account quota, please retry later type: api_error code: insufficient_user_quota param: '' '500': description: Internal server error while processing the request. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' examples: internal_error: summary: Internal processing failure value: error: message: Internal processing failed, please retry later type: api_error code: gen_relay_info_failed param: '' security: - BearerAuth: [] components: schemas: ChatCompletionRequest: type: object required: - model - messages description: MIMO chat completions request body. properties: model: type: string description: Model ID to use enum: - mimo-v2.5 example: mimo-v2.5 messages: type: array description: A list of messages comprising the conversation so far. minItems: 1 items: $ref: '#/components/schemas/Message' frequency_penalty: type: number nullable: true default: 0 minimum: -2 maximum: 2 description: >- A value between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, reducing the model's likelihood of repeating the same content. presence_penalty: type: number nullable: true default: 0 minimum: -2 maximum: 2 description: >- Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they have appeared in the text so far, increasing the model's likelihood to talk about new topics. max_completion_tokens: type: integer nullable: true description: >- The upper limit of tokens that can be generated in the chat completion, including visible output tokens and reasoning tokens. Default is 32768 for mimo-v2.5. Range: [1, 131072] response_format: $ref: '#/components/schemas/ResponseFormat' stop: description: >- Up to 4 sequences where the API will stop generating further tokens when any of them is encountered. The returned text will not contain the stop sequences. nullable: true default: null oneOf: - type: string - type: array items: type: string maxItems: 4 stream: type: boolean nullable: true default: false description: >- If set to true, the model's response data will be streamed to the client via SSE (server-sent events) during generation. thinking: $ref: '#/components/schemas/ThinkingConfig' temperature: type: number minimum: 0 maximum: 1.5 default: 1 description: >- Sampling temperature to use, between 0 and 1.5. Higher values (e.g., 0.8) make the output more random, while lower values (e.g., 0.2) make it more focused and deterministic. We generally recommend altering this or top_p but not both. In thinking mode, mimo-v2.5-pro and mimo-v2.5 do not support custom temperature. Even if passed, the model will force the recommended default value of 1.0. tool_choice: $ref: '#/components/schemas/ToolChoice' top_p: type: number default: 0.95 minimum: 0.01 maximum: 1 description: >- Probability threshold for nucleus sampling, controlling the diversity of text generation. Higher top_p values produce more diverse text; lower values produce more deterministic text. Since both temperature and top_p control text diversity, it is recommended to adjust only one of them. In thinking mode, mimo-v2.5-pro and mimo-v2.5 do not support custom top_p. Even if passed, the model will force the recommended default value of 0.95. tools: type: array description: >- A list of tools the model may call. Currently, both function tools and web_search tools are supported. Note: In multi-round tool calls under thinking mode, the model returns the reasoning_content field along with tool_calls. To continue the conversation, it is recommended to retain all historical reasoning_content in the messages array of each subsequent request for optimal performance. items: $ref: '#/components/schemas/ToolDefinition' ChatCompletionResponse: type: object description: Chat response object (non-streaming output). properties: id: type: string description: A unique identifier for the response. object: type: string description: The type of the object, which is chat.completion. created: type: integer description: >- The Unix timestamp (in seconds) when the chat completion was created. model: type: string description: The name of the model that generated the completion. choices: type: array description: A list of completions generated by the model. items: $ref: '#/components/schemas/ChatCompletionChoice' usage: $ref: '#/components/schemas/Usage' ChatCompletionChunk: type: object description: >- Chat response chunk object (streaming output). Each `data:` event in the SSE stream corresponds to a ChatCompletionChunk. The stream typically ends with `data: [DONE]`. properties: id: type: string description: >- Unique identifier for the chat completion object. Each data block uses the same ID. object: type: string enum: - chat.completion.chunk description: The object type, which is always chat.completion.chunk. created: type: integer description: >- Unix timestamp (in seconds) when the chat completion object was created. Each data block uses the same timestamp. model: type: string description: The model used to generate the response. choices: type: array description: List of generated reply options. items: $ref: '#/components/schemas/ChatCompletionChunkChoice' usage: description: Usage information for the chat completion request. allOf: - $ref: '#/components/schemas/Usage' nullable: true ErrorResponse: type: object properties: error: $ref: '#/components/schemas/ErrorObject' Message: description: >- A single message in the conversation. Depending on the role, it is divided into five message types: developer, system, user, assistant, and tool. oneOf: - $ref: '#/components/schemas/DeveloperMessage' - $ref: '#/components/schemas/SystemMessage' - $ref: '#/components/schemas/UserMessage' - $ref: '#/components/schemas/AssistantRequestMessage' - $ref: '#/components/schemas/ToolMessage' ResponseFormat: type: object description: An object specifying the format that the model must output. oneOf: - type: object title: Text description: Default response format. Used for generating text responses. properties: type: type: string enum: - text description: The type of response format. Must be text. required: - type - type: object title: JSON object description: >- JSON object response format. Note: The model will not output JSON formatted content unless instructed to generate JSON in the system or user messages. properties: type: type: string enum: - json_object description: The type of response format. Must be json_object. required: - type ThinkingConfig: type: object description: >- Controls whether the model enables chain-of-thought (thinking) mode. Note: In multi-round tool calls under thinking mode, the model returns the reasoning_content field along with tool_calls. To continue the conversation, it is recommended to retain all historical reasoning_content in the messages array of each subsequent request for optimal performance. In thinking mode, mimo-v2.5-pro and mimo-v2.5 do not support custom temperature and top_p parameters. Even if passed, the model will force the recommended default values of 1.0 and 0.95 respectively. properties: type: type: string enum: - enabled - disabled default: enabled description: >- Whether to enable chain-of-thought. For mimo-v2.5-pro and mimo-v2.5, the default value is enabled. Allowed values: enabled, disabled required: - type ToolChoice: oneOf: - type: string enum: - auto description: >- Controls how the model selects tools. Note: When tool_choice is set to a non-auto value, the backend will remove this field by default, and the model's response behavior will still be equivalent to auto mode (this logic may be subject to adjustment). Allowed values: auto ToolDefinition: oneOf: - type: object title: Function tool required: - type - function description: Function tool definition. properties: type: type: string enum: - function description: The type of the tool. Currently function is supported. function: type: object required: - name properties: name: type: string description: >- The name of the function to call. Must consist of a-z, A-Z, 0-9 characters, or include underscores (_) and hyphens (-), with a maximum length of 64 characters. description: type: string description: >- A description of what the function does, used by the model to understand when and how to call it. parameters: type: object description: >- The input parameters of the function, described as a JSON Schema object. Omitting parameters defines a function with an empty parameter list. additionalProperties: true strict: type: boolean default: false description: >- Whether to enable strict schema adherence when generating function calls. If set to true, the model will strictly follow the exact schema defined in the parameters field. When strict is true, only a subset of JSON Schema is supported. - $ref: '#/components/schemas/WebSearchTool' ChatCompletionChoice: type: object properties: index: type: integer description: The index of this completion in the list of choices. message: $ref: '#/components/schemas/AssistantResponseMessage' finish_reason: type: string description: >- The reason the model stopped generating tokens. stop if the model reached a natural stop point or a provided stop sequence; length if the maximum number of tokens specified in the request was reached; tool_calls if the model called a tool; content_filter if the content was blocked due to a filtering policy; repetition_truncation if the model detected repetition. Usage: type: object description: Usage information for this chat completion request. properties: prompt_tokens: type: integer description: Number of tokens used by the prompt. completion_tokens: type: integer description: Number of tokens used by the model output. total_tokens: type: integer description: Total number of tokens used in the request (prompt + completion). completion_tokens_details: type: object description: Breakdown of tokens used in the completion. properties: reasoning_tokens: type: integer description: Number of tokens generated by the model for reasoning. prompt_tokens_details: type: object description: Breakdown of tokens used in the prompt. properties: cached_tokens: type: integer description: Number of tokens that hit the cache. audio_tokens: type: integer description: Number of audio input tokens present in the prompt. image_tokens: type: integer description: Number of image input tokens present in the prompt. video_tokens: type: integer description: Number of video input tokens present in the prompt. web_search_usage: type: object description: Usage details for the web search API. properties: tool_usage: type: integer description: Number of web search API calls. page_usage: type: integer description: Number of web pages returned by the web search API. ChatCompletionChunkChoice: type: object description: Chat response chunk object (streaming). properties: index: type: integer description: The index of this completion in the list of choices. delta: type: object description: The chat completion delta generated by the model in streaming mode. properties: role: type: string description: The role of the message author. content: type: string description: The content of the chunk message. reasoning_content: type: string description: >- The reasoning content in the assistant message, preceding the final answer. tool_calls: type: array description: >- After the function call is initiated, the model returns the tools to be called and the parameters required for the call. This parameter may contain one or more tool response objects. items: type: object properties: index: type: integer description: >- The index of the tool call in the tool_calls list, starting from 0. id: type: string description: The ID of the tool call. type: type: string enum: - function description: >- The type of the tool. Currently only function is supported. function: type: object description: The function being called. properties: name: type: string description: The name of the function to call. arguments: type: string description: >- Model-generated arguments for calling the function, formatted as JSON. Note that the model-generated content is not always guaranteed to be valid JSON and may fabricate parameters not defined in the function schema. Before calling the function, validate these parameters in your code. annotations: type: array description: >- After web search, the model returns annotations for all cited URLs. items: $ref: '#/components/schemas/Annotation' error_message: type: string description: Error message for the web search chain. finish_reason: type: string nullable: true description: >- The reason the model stopped generating tokens. stop if the model reached a natural stop point or a provided stop sequence; length if the maximum number of tokens specified in the request was reached; tool_calls if the model called a tool; content_filter if the content was blocked due to a filtering policy; repetition_truncation if the model detected repetition. ErrorObject: type: object properties: message: type: string description: Error description. type: type: string description: Error type. param: type: string description: Associated field name. code: type: string description: Error code. DeveloperMessage: type: object title: Developer message · object required: - role - content description: >- Developer message. Instructions provided by the developer that the model should follow, regardless of the messages sent by the user. properties: role: type: string enum: - developer description: The role of the message author, which is developer. content: description: The content of the developer message. oneOf: - type: string description: Plain text content. - $ref: '#/components/schemas/TextContentPartArray' name: type: string description: >- An optional name for the participant, provided to the model to distinguish between participants of the same role. SystemMessage: title: System message · object type: object required: - role - content description: System message. The role of this message is system. properties: content: description: The content of the system message. oneOf: - type: string description: Plain text content. - $ref: '#/components/schemas/TextContentPartArray' role: type: string enum: - system description: The role of the message author, which is system. name: type: string description: >- An optional name for the participant, provided to the model to distinguish between participants of the same role. UserMessage: title: User message · object type: object required: - role - content description: User message. The role of this message is user. properties: content: description: The content of the user message. oneOf: - type: string description: Plain text content. - $ref: '#/components/schemas/ContentPartArray' role: type: string enum: - user description: The role of the message author, which is user. name: type: string description: >- An optional name for the participant, provided to the model to distinguish between participants of the same role. AssistantRequestMessage: title: Assistant message · object type: object required: - role description: Assistant message. The role of this message is assistant. properties: content: description: >- The content of the assistant message. Required unless tool_calls is specified. oneOf: - type: string nullable: true description: Plain text content. - $ref: '#/components/schemas/TextContentPartArray' role: type: string enum: - assistant description: The role of the message author, which is assistant. name: type: string description: >- An optional name for the participant, provided to the model to distinguish between participants of the same role. tool_calls: type: array description: The tool calls generated by the model, such as function calls. items: $ref: '#/components/schemas/ToolCallMessage' ToolMessage: title: Tool message · object type: object required: - role - content - tool_call_id description: Tool message. The role of this message is tool. properties: role: type: string enum: - tool description: The role of the message author, which is tool. content: description: The content of the tool message. oneOf: - type: string description: Plain text content. - $ref: '#/components/schemas/TextContentPartArray' tool_call_id: type: string description: The ID of the tool call that this message is responding to. WebSearchTool: type: object required: - type description: >- Web search tool object. For details, please refer to the web search documentation. Note: Web search tools are not supported at this time. properties: user_location: $ref: '#/components/schemas/WebSearchUserLocation' type: type: string enum: - web_search description: The type of the tool. Must be web_search. force_search: type: string default: 'false' description: >- Whether to force search. true forces search, false lets the model decide whether search is needed. max_keyword: type: integer default: 5 minimum: 1 maximum: 50 description: >- Limits the maximum number of keywords that can be used in a single search turn. limit: type: integer default: 5 minimum: 1 maximum: 50 description: >- Limits the maximum number of results returned per single search operation. title: Web search tool AssistantResponseMessage: type: object description: The chat completion message generated by the model. properties: role: type: string description: The role of the message author. content: type: string description: The content of the message. reasoning_content: type: string description: >- The reasoning content in the assistant message, preceding the final answer. tool_calls: type: array description: >- After a function call is initiated, the model returns the tools to be called along with the parameters required. This parameter may contain one or more tool response objects. items: oneOf: - $ref: '#/components/schemas/ToolCallMessage' annotations: type: array description: After web search, the model returns annotations for all cited URLs. items: $ref: '#/components/schemas/Annotation' error_message: type: string description: Error message for web search. audio: $ref: '#/components/schemas/AudioResponse' final_text_preview: type: string description: >- The final text after intelligent optimization and polishing for audio broadcasting. Only returned when the request parameter optimize_text_preview is set to true. Annotation: type: object description: Annotations returned by web search, containing citation URLs. properties: logo_url: type: string description: Logo URL. publish_time: type: string description: Publish time. site_name: type: string description: Website name. summary: type: string description: Summary. title: type: string description: Title. type: type: string description: Type. url: type: string description: URL. TextContentPartArray: type: array description: An array of plain text content parts within a message. items: $ref: '#/components/schemas/TextContentPart' title: Array of content parts · array ContentPartArray: type: array description: >- An array of content parts in a user message. May include text, images, audio, or video. items: oneOf: - $ref: '#/components/schemas/ContentPartText' - $ref: '#/components/schemas/ContentPartImage' - $ref: '#/components/schemas/ContentPartAudio' - $ref: '#/components/schemas/ContentPartVideo' ToolCallMessage: type: object description: Tool call object within a message. properties: id: type: string description: The ID of the tool call. type: type: string enum: - function description: The type of the tool. Currently only function is supported. function: type: object description: The function the model called. properties: name: type: string description: The name of the function the model called. arguments: type: string description: >- The arguments to call the function with, generated by the model as a JSON string. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined in your function schema. Validate these arguments in your code before calling the function. title: Function tool call · object WebSearchUserLocation: type: object required: - type - country description: User geographical location for optimizing model output. properties: type: type: string enum: - approximate description: Location type, must be approximate. country: type: string description: Country region: type: string description: Region/State city: type: string description: City district: type: string description: District longitude: type: number format: double description: Longitude latitude: type: number format: double description: Latitude AudioResponse: type: object description: >- If audio output is requested, this object contains data about the model's audio response. properties: id: type: string description: Unique identifier for this audio response. data: type: string description: >- Base64-encoded audio generated by the model, in the format specified in the request. expires_at: type: number nullable: true description: >- Unix timestamp (in seconds) when this audio response expires. Currently null. transcript: type: string nullable: true description: Transcript of the generated audio. Currently null. TextContentPart: type: object title: Text content part · object required: - type - text description: A plain text content part within a message. properties: type: type: string enum: - text description: The type of the content part, which is text. text: type: string description: The text content. ContentPartText: type: object title: Text content part · object required: - type - text description: A plain text content part within a user message. properties: type: type: string enum: - text description: The type of the content part, which is text. text: type: string description: The text content. ContentPartImage: type: object title: Image content part · object required: - type - image_url description: An image content part within a user message. properties: type: type: string enum: - image_url description: The type of the content part, which is image_url. image_url: type: object required: - url description: >- The image URL. May be a public image URL or a Base64-encoded data URL. properties: url: type: string format: uri description: A public image URL or a Base64-encoded data URL. ContentPartAudio: type: object title: Audio content part · object required: - type - input_audio description: An audio content part within a user message. properties: type: type: string enum: - input_audio description: The type of the content part, which is input_audio. input_audio: type: object required: - data description: The audio URL or Base64-encoded audio data. properties: data: type: string description: The URL of the audio or Base64-encoded audio data. ContentPartVideo: type: object title: Video content part · object required: - type - video_url description: A video content part within a user message. properties: type: type: string enum: - video_url description: The type of the content part, which is video_url. video_url: type: object required: - url description: >- The video URL. May be a public video URL or a Base64-encoded data URL. properties: url: type: string format: uri description: The URL of the video or Base64-encoded video data. fps: type: number default: 2 description: >- The number of frames to extract per second. Valid range: [0.1, 10.0] media_resolution: type: string default: default enum: - default - max description: 'Resolution level. Allowed values: default, max' securitySchemes: BearerAuth: type: http scheme: bearer bearerFormat: API Token description: 'Pass `Authorization: Bearer ` in the request header.' ````