Reka

aid: reka
name: Reka
description: >-
  Reka is a multimodal foundation model company building natively multimodal
  large language models capable of joint reasoning over text, image, video,
  and audio. The Reka model family spans Spark (1B) for embedded use cases,
  Edge (7B), Flash (21B), and Core (67B) for complex enterprise tasks. Reka
  exposes its hosted models through the Reka API at api.reka.ai with
  OpenAI-compatible chat completions, and also ships products including the
  Reka Vision Platform for enterprise perception, Reka Speech for audio
  understanding, and Reka Clip for creators. The platform supports flexible
  deployment across cloud, VPC, on-premises, and fully air-gapped
  environments, and the company maintains an active open-source presence on
  Hugging Face and GitHub including the Reka VibeEval benchmark.
type: Index
position: Provider
access: 3rd-Party
image: https://kinlane-productions.s3.amazonaws.com/apis-json/apis-json-logo.jpg
tags:
  - AI
  - Multimodal
  - Large Language Models
  - Vision
  - Speech
  - Foundation Models
  - OpenAI Compatible
  - SDK
  - Enterprise
  - On-Premises
url: https://raw.githubusercontent.com/api-evangelist/reka/refs/heads/main/apis.yml
created: '2026-05-23'
modified: '2026-05-23'
specificationVersion: '0.20'
apis:
  - aid: reka:reka-api
    name: Reka API
    description: >-
      The Reka API exposes chat-completion and multimodal endpoints for the
      Reka Core, Flash, Edge, and Spark models with image, video, and audio
      input support. The v1 surface at api.reka.ai/v1 is OpenAI-compatible
      and Reka Research is fully compatible with the OpenAI /chat/completions
      schema. Reka offers a published OpenAPI definition and interactive
      Swagger documentation along with first-party Python and TypeScript
      SDKs, vLLM plugin, n8n nodes, and a .NET SDK.
    humanURL: https://docs.reka.ai
    baseURL: https://api.reka.ai/v1
    tags:
      - Multimodal
      - LLM
      - Chat Completions
      - OpenAI Compatible
      - Vision
      - Video
      - REST
    properties:
      - type: Documentation
        url: https://docs.reka.ai
      - type: GettingStarted
        url: https://docs.reka.ai/quick-start
      - type: OpenAPI
        url: https://api.reka.ai/openapi.json
      - type: Developer Portal
        url: https://platform.reka.ai
      - type: SDK
        url: https://github.com/reka-ai/reka-sdk-python
      - type: SDK
        url: https://pypi.org/project/reka-api
      - type: Benchmark
        url: https://github.com/reka-ai/reka-vibe-eval
    features:
      - name: Multimodal Chat Completions
        description: >-
          Joint reasoning over text, image, video, and audio inputs in a
          single chat-completion request.
      - name: Tiered Model Catalog
        description: >-
          Core (67B), Flash (21B), Edge (7B), and Spark (1B) models for
          different cost, latency, and capability points.
      - name: OpenAI Compatibility
        description: >-
          Reka Research is fully compatible with the OpenAI
          /chat/completions schema for drop-in client use.
      - name: Vision Platform
        description: >-
          Enterprise vision platform for detection, captioning, and
          video understanding workloads.
      - name: Speech Understanding
        description: >-
          Reka Speech endpoint for audio comprehension and transcription
          with multimodal reasoning.
      - name: Flexible Deployment
        description: >-
          Cloud, VPC, on-premises, and air-gapped deployment options for
          regulated and sovereign customers.
      - name: First-Party SDKs
        description: >-
          Python, TypeScript, and .NET SDKs with examples, plus a vLLM
          plugin for self-hosted Reka models.
    useCases:
      - name: Video Captioning and Search
        description: >-
          Generate captions, summaries, and chapter markers for long-form
          video content.
      - name: Security and Defense Perception
        description: >-
          Detection, classification, and event reasoning over surveillance
          and operations video.
      - name: Sports and Game Highlights
        description: >-
          Auto-generate highlight reels and performance analytics from
          live and recorded footage.
      - name: Enterprise Document Understanding
        description: >-
          Reason over multimodal documents combining text, tables, and
          embedded imagery.
      - name: On-Premises Sovereign AI
        description: >-
          Deploy Reka models inside customer VPC, on-prem, or air-gapped
          environments for regulated workloads.
    integrations:
      - name: OpenAI SDK
      - name: LangChain
      - name: LlamaIndex
      - name: Hugging Face
      - name: vLLM
      - name: n8n
      - name: Vercel AI SDK
      - name: Discord
    authentication:
      - type: API Key
        description: >-
          Authenticate with a REKA_API_KEY Bearer token issued from the
          Reka Platform dashboard at platform.reka.ai.
common:
  - type: Website
    url: https://reka.ai
  - type: Documentation
    url: https://docs.reka.ai
  - type: Developer Portal
    url: https://platform.reka.ai
  - type: OpenAPI
    url: https://api.reka.ai/openapi.json
  - type: GitHubOrganization
    url: https://github.com/reka-ai
  - type: HuggingFace
    url: https://huggingface.co/RekaAI
  - type: Discord
    url: https://discord.gg/rekaai
  - type: Blog
    url: https://reka.ai/news
  - type: LinkedIn
    url: https://www.linkedin.com/company/reka-ai
  - type: Twitter
    url: https://twitter.com/RekaAILabs
maintainers:
  - FN: Kin Lane
    email: [email protected]