> ## Documentation Index > Fetch the complete documentation index at: https://docs.deepinfra.com/llms.txt > Use this file to discover all available pages before exploring further. # Openai Completions ## OpenAPI ````yaml https://api.deepinfra.com/openapi.json post /v1/completions openapi: 3.1.0 info: title: DeepInfra API description: >- The DeepInfra API provides serverless AI inference, custom model deployments, and GPU rentals. version: 1.0.0 servers: - url: https://api.deepinfra.com security: [] tags: - name: Chat Completions description: OpenAI and Anthropic-compatible chat completion endpoints for LLMs. - name: Text Completions description: OpenAI-compatible text completion endpoints. - name: Embeddings description: Generate text embeddings for search and RAG. - name: Image Generation description: Generate, edit, and create variations of images. - name: Audio description: OpenAI-compatible speech synthesis, transcription, and translation. - name: Text to Speech description: ElevenLabs-compatible TTS endpoints and voice management. - name: Inference description: Native DeepInfra inference API for models and deployments. - name: Dedicated Models description: Deploy and manage private model instances with autoscaling. - name: GPU Rentals description: Rent dedicated GPU containers. - name: Models description: Browse, search, and manage AI models. - name: Files & Batches description: File uploads and batch processing. - name: LoRA Adapters description: Create, manage, and query LoRA adapter models. - name: Agents description: Manage agent-framework instances (OpenClaw and friends). - name: Sandboxes description: Create and manage isolated sandbox environments. - name: Account description: User profile, team management, and rate limits. - name: Authentication description: API tokens, SSH keys, scoped JWTs, and login flows. - name: Billing description: Payment methods, usage tracking, and billing. - name: Logs & Metrics description: Query inference logs, deployment logs, and usage metrics. - name: Utilities description: Feedback submission and CLI version. paths: /v1/completions: post: tags: - Text Completions summary: Openai Completions operationId: openai_completions_v1_completions_post parameters: - name: x-deepinfra-source in: header required: false schema: anyOf: - type: string - type: 'null' title: X-Deepinfra-Source - name: xi-api-key in: header required: false schema: anyOf: - type: string - type: 'null' title: Xi-Api-Key - name: x-api-key in: header required: false schema: anyOf: - type: string - type: 'null' title: X-Api-Key requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/OpenAICompletionsIn' responses: '200': description: Successful Response content: application/json: schema: {} '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' security: - HTTPBearer: [] components: schemas: OpenAICompletionsIn: properties: service_tier: anyOf: - $ref: '#/components/schemas/ServiceTier' - type: 'null' description: >- The service tier used for processing the request. 'priority' processes the request with higher priority (premium rate); 'flex' processes it at lower priority for a discount, served only when spare capacity exists and may be retried/timed out under load. Both apply only to models that support the respective tier. model: type: string title: Model description: model name examples: - meta-llama/Llama-2-70b-chat-hf prompt: anyOf: - type: string - items: type: integer type: array title: Prompt description: input prompt - a single string is currently supported max_tokens: anyOf: - type: integer maximum: 10000000 exclusiveMinimum: 0 - type: 'null' title: Max Tokens description: >- The maximum number of tokens to generate in the completion. The total length of input tokens and generated tokens is limited by the model's context length.If explicitly set to None it will be the model's max context length minus input length or 65536, whichever is smaller. temperature: type: number maximum: 2 minimum: 0 title: Temperature description: >- What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic default: 1 top_p: type: number maximum: 1 exclusiveMinimum: 0 title: Top P description: >- An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. default: 1 min_p: type: number maximum: 1 minimum: 0 title: Min P description: >- Float that represents the minimum probability for a token to be considered, relative to the probability of the most likely token. Must be in [0, 1]. Set to 0 to disable this. default: 0 top_k: type: integer exclusiveMaximum: 1000 minimum: 0 title: Top K description: Sample from the best k (number of) tokens. 0 means off default: 0 'n': type: integer maximum: 4 minimum: 1 title: 'N' description: number of sequences to return default: 1 stream: type: boolean title: Stream description: whether to stream the output via SSE or return the full response default: false logprobs: anyOf: - type: integer - type: 'null' title: Logprobs description: return top tokens and their log-probabilities echo: anyOf: - type: boolean - type: 'null' title: Echo description: return prompt as part of the respons stop: anyOf: - type: string - items: type: string type: array - type: 'null' title: Stop description: up to 16 sequences where the API will stop generating further tokens presence_penalty: type: number maximum: 2 minimum: -2 title: Presence Penalty description: >- Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics. default: 0 frequency_penalty: type: number maximum: 2 minimum: -2 title: Frequency Penalty description: >- Positive values penalize new tokens based on how many times they appear in the text so far, increasing the model's likelihood to talk about new topics. default: 0 response_format: anyOf: - $ref: '#/components/schemas/TextResponseFormat' - $ref: '#/components/schemas/JsonObjectResponseFormat' - $ref: '#/components/schemas/JsonSchemaResponseFormat' - $ref: '#/components/schemas/RegexResponseFormat' - type: 'null' title: Response Format description: The format of the response. Currently, only json is supported. repetition_penalty: type: number maximum: 5 minimum: 0.01 title: Repetition Penalty description: >- Alternative penalty for repetition, but multiplicative instead of additive (> 1 penalize, < 1 encourage) default: 1 user: anyOf: - type: string - type: 'null' title: User description: >- A unique identifier representing your end-user, which can help monitor and detect abuse. Avoid sending us any identifying information. We recommend hashing user identifiers. seed: anyOf: - type: integer exclusiveMaximum: 18446744073709552000 minimum: -9223372036854776000 - type: 'null' title: Seed description: >- Seed for random number generator. If not provided, a random seed is used. Determinism is not guaranteed. stream_options: anyOf: - $ref: '#/components/schemas/StreamOptions' - type: 'null' description: streaming options stop_token_ids: anyOf: - items: type: integer type: array - type: 'null' title: Stop Token Ids description: >- Up to 16 token IDs where the API will stop generating further tokens. Merged with the model's built-in stop tokens. Intended for private deployments. return_tokens_as_token_ids: anyOf: - type: boolean - type: 'null' title: Return Tokens As Token Ids description: return tokens as token ids prompt_cache_key: anyOf: - type: string - type: 'null' title: Prompt Cache Key description: >- A key to identify prompt cache for reuse across requests. If provided, the prompt will be cached and can be reused in subsequent requests with the same key. data: anyOf: - $ref: '#/components/schemas/CompletionMultiModalData' - type: 'null' description: >- Optional multi-modal data to pass alongside the prompt. Only supported for a small number of non-chat-native vision models. Images must be base64 data URIs (e.g. 'data:image/png;base64,...'). type: object required: - model - prompt title: OpenAICompletionsIn HTTPValidationError: properties: detail: items: $ref: '#/components/schemas/ValidationError' type: array title: Detail type: object title: HTTPValidationError ServiceTier: type: string enum: - default - priority - flex title: ServiceTier TextResponseFormat: properties: type: type: string const: text title: Type default: text type: object title: TextResponseFormat JsonObjectResponseFormat: properties: type: type: string const: json_object title: Type default: json_object type: object title: JsonObjectResponseFormat JsonSchemaResponseFormat: properties: type: type: string const: json_schema title: Type default: json_schema json_schema: $ref: '#/components/schemas/JsonSchema' description: JSON schema for structured output when type is 'json_schema' type: object required: - json_schema title: JsonSchemaResponseFormat RegexResponseFormat: properties: type: type: string const: regex title: Type default: regex regex: type: string title: Regex description: Regex pattern for structured output when type is 'regex' type: object required: - regex title: RegexResponseFormat StreamOptions: properties: include_usage: type: boolean title: Include Usage description: whether to include usage data default: true continuous_usage_stats: type: boolean title: Continuous Usage Stats description: >- whether to include usage stats continuously with each streaming event default: false type: object title: StreamOptions CompletionMultiModalData: properties: image: items: type: string type: array title: Image description: >- List of images as base64 data URIs (e.g. 'data:image/png;base64,...'). Each image must correspond to a placeholder token in the prompt. video: items: type: string type: array title: Video description: >- List of videos as base64 data URIs (e.g. 'data:video/mp4;base64,...'). Each video must correspond to a placeholder token in the prompt. type: object title: CompletionMultiModalData ValidationError: properties: loc: items: anyOf: - type: string - type: integer type: array title: Location msg: type: string title: Message type: type: string title: Error Type type: object required: - loc - msg - type title: ValidationError JsonSchema: properties: name: type: string title: Name description: Name identifier for the JSON schema schema: additionalProperties: true type: object title: Schema description: The actual JSON schema definition type: object required: - name - schema title: JsonSchema securitySchemes: HTTPBearer: type: http scheme: bearer ````