tri.ai.openai.resources.openai-models.yaml Maven / Gradle / Ivy

Go to download
Show more of this group Show more artifacts with this name
Show all versions of promptkt Show documentation
LLM and prompt engineering.
The newest version!
# Library of OpenAI Models
# See https://platform.openai.com/docs/models
---
models:

  GPT-4:

    - id: gpt-4-turbo
      name: GPT-4-turbo
      type: TEXT_VISION_CHAT
      source: OpenAI
      description: |
        The latest GPT-4 Turbo model with vision capabilities. Vision requests can now use JSON mode and function calling.
      totalTokenLimit: 128000
      snapshots: [ '2024-04-09', 'preview' ]

    - id: gpt-4
      name: GPT-4
      type: TEXT_CHAT
      source: OpenAI
      totalTokenLimit: 8192
      snapshots: [ '0613', '1106-preview' ]

  GPT-4 multimodal:

    - id: gpt-4o
      name: GPT-4o
      type: TEXT_VISION_CHAT
      source: OpenAI
      description: |
        GPT-4o (“o” for “omni”) is our most advanced model. It is multimodal (accepting text or image inputs and outputting text), 
        and it has the same high intelligence as GPT-4 Turbo but is much more efficient—it generates text 2x faster and is 50% cheaper. 
        Additionally, GPT-4o has the best vision and performance across non-English languages of any of our models.
      totalTokenLimit: 128000
      snapshots: [ '2024-05-13' ]

    - id: gpt-4-vision-preview
      name: GPT-4 Turbo with Vision Preview
      type: TEXT_VISION_CHAT
      source: OpenAI
      description: |
        GPT-4 with ability to understand images.
        Returns a maximum of 4,096 output tokens.
      totalTokenLimit: 128000

  GPT-3.5:

    - id: gpt-3.5-turbo
      name: GPT-3.5 Turbo
      type: TEXT_CHAT
      source: OpenAI
      description: |
        GPT-3.5 with improved instruction following, JSON mode, reproducible outputs, parallel function calling, and more.
        Returns a maximum of 4,096 output tokens.
      totalTokenLimit: 16385
      snapshots: [ '1106', '0125' ]

  Embeddings:

    - id: text-embedding-ada-002
      name: Ada Text Embeddings
      type: TEXT_EMBEDDING
      source: OpenAI
      description: Most capable 2nd generation embedding model, replacing 16 first generation models
      outputDimension: 1536

    - id: text-embedding-3-large
      name: Ada Text Embeddings
      type: TEXT_EMBEDDING
      source: OpenAI
      description: Most capable embedding model for both english and non-english tasks
      outputDimension: 3072

    - id: text-embedding-3-small
      name: Ada Text Embeddings
      type: TEXT_EMBEDDING
      source: OpenAI
      description: Increased performance over 2nd generation ada embedding model
      outputDimension: 1536

  GPT completions:

    - id: gpt-3.5-turbo-instruct
      name: GPT-3.5 Turbo Instruct
      type: TEXT_COMPLETION
      source: OpenAI
      totalTokenLimit: 4096
      snapshots: [ ]

    - id: davinci-002
      name: Davinci 2
      type: TEXT_COMPLETION
      source: OpenAI
      totalTokenLimit: 16384

    - id: babbage-002
      name: Babbage 2
      type: TEXT_COMPLETION
      source: OpenAI
      totalTokenLimit: 16384

  GPT-3 legacy:

    - id: text-davinci-edit-001
      name: DaVinci text edit
      type: UNKNOWN
      source: OpenAI
      deprecation: 2024-01-04

    - id: code-davinci-edit-001
      name: DaVinci code edit
      type: UNKNOWN
      source: OpenAI
      deprecation: 2024-01-04

  Image Generation:

    - id: dall-e-3
      name: DALL-E 3
      type: IMAGE_GENERATOR
      source: OpenAI

    - id: dall-e-2
      name: DALL-E 2
      type: IMAGE_GENERATOR
      source: OpenAI

  Moderation:

    - id: text-moderation-latest
      name: Text Moderation (latest)
      type: MODERATION
      source: OpenAI
      totalTokenLimit: 32768

    - id: text-moderation-stable
      name: Text Moderation (stable)
      type: MODERATION
      source: OpenAI
      totalTokenLimit: 32768

  Speech:

    - id: tts-1
      name: TTS 1
      source: OpenAI
      type: TEXT_TO_SPEECH
      description: TTS 1 optimized for real-time text
      snapshots: [ '1106' ]

    - id: tts-1-hd
      name: TTS 1 HD
      type: TEXT_TO_SPEECH
      source: OpenAI
      description: TTS 1 optimized for higher quality audio
      snapshots: [ '1106' ]

    - id: canary-tts
      name: Canary TTS
      type: TEXT_TO_SPEECH
      source: OpenAI
      description: Canary TTS

    - id: whisper-1
      name: Whisper v2-large
      type: SPEECH_TO_TEXT
      source: OpenAI
      description: Whisper speech recognition model, see https://github.com/openai/whisper and https://arxiv.org/abs/2212.04356

    - id: canary-whisper
      name: Canary Whisper
      type: SPEECH_TO_TEXT
      source: OpenAI
      description: Canary Whisper

audio: [ whisper-1 ]
chat: [
  gpt-3.5-turbo, gpt-4o, gpt-4-turbo, gpt-4
]
completion: [ gpt-3.5-turbo-instruct, davinci-002, babbage-002 ]
embeddings: [ text-embedding-ada-002, text-embedding-3-large, text-embedding-3-small ]
moderation: [ text-moderation-latest, text-moderation-stable ]
tts: [ tts-1, tts-1-hd ]
image_generator: [ dall-e-2, dall-e-3 ]
vision_language: [ gpt-4-vision-preview, gpt-4o ]