AWS Bedrock - Supported Models

Supported AWS Bedrock Models

Titan Text Express V1 with id amazon.titan-text-express-v1 Express is a large language model for text generation. It is useful for a wide range of advanced, general language tasks such as open-ended text generation and conversational chat, as well as support within Retrieval Augmented Generation (RAG).

Json schema for request

{
  "$schema": "http://json-schema.org/draft-04/schema#",
  "type": "object",
  "properties": {
    "inputText": {
      "type": "string"
    },
    "textGenerationConfig": {
      "type": "object",
      "properties": {
        "maxTokenCount": {
          "type": "integer"
        },
        "stopSequences": {
          "type": "array",
          "items": [
            {
              "type": "string"
            }
          ]
        },
        "temperature": {
          "type": "integer"
        },
        "topP": {
          "type": "integer"
        }
      },
      "required": [
        "maxTokenCount",
        "stopSequences",
        "temperature",
        "topP"
      ]
    }
  },
  "required": [
    "inputText",
    "textGenerationConfig"
  ]
}

Titan Text Lite V1 with id amazon.titan-text-lite-v1 Lite is a light weight efficient model, ideal for fine-tuning of English-language tasks.

Json schema for request

{
  "$schema": "http://json-schema.org/draft-04/schema#",
  "type": "object",
  "properties": {
    "inputText": {
      "type": "string"
    },
    "textGenerationConfig": {
      "type": "object",
      "properties": {
        "maxTokenCount": {
          "type": "integer"
        },
        "stopSequences": {
          "type": "array",
          "items": [
            {
              "type": "string"
            }
          ]
        },
        "temperature": {
          "type": "integer"
        },
        "topP": {
          "type": "integer"
        }
      },
      "required": [
        "maxTokenCount",
        "stopSequences",
        "temperature",
        "topP"
      ]
    }
  },
  "required": [
    "inputText",
    "textGenerationConfig"
  ]
}

Titan Image Generator G1 with id amazon.titan-image-generator-v1 It generates images from text, and allows users to upload and edit an existing image. Users can edit an image with a text prompt (without a mask) or parts of an image with an image mask. You can extend the boundaries of an image with outpainting, and fill in an image with inpainting.

Json schema for request

{
  "$schema": "http://json-schema.org/draft-04/schema#",
  "type": "object",
  "properties": {
    "textToImageParams": {
      "type": "object",
      "properties": {
        "text": {
          "type": "string"
        },
        "negativeText": {
          "type": "string"
        }
      },
      "required": [
        "text",
        "negativeText"
      ]
    },
    "taskType": {
      "type": "string"
    },
    "imageGenerationConfig": {
      "type": "object",
      "properties": {
        "cfgScale": {
          "type": "integer"
        },
        "seed": {
          "type": "integer"
        },
        "quality": {
          "type": "string"
        },
        "width": {
          "type": "integer"
        },
        "height": {
          "type": "integer"
        },
        "numberOfImages": {
          "type": "integer"
        }
      },
      "required": [
        "cfgScale",
        "seed",
        "quality",
        "width",
        "height",
        "numberOfImages"
      ]
    }
  },
  "required": [
    "textToImageParams",
    "taskType",
    "imageGenerationConfig"
  ]
}

Titan Embeddings G1 with id amazon.titan-embed-text-v1 The Amazon Titan Embeddings G1 - Text – Text v1.2 can intake up to 8k tokens and outputs a vector of 1,536 dimensions. The model also works in 25+ different language

Json schema for request

{
  "$schema": "http://json-schema.org/draft-04/schema#",
  "type": "object",
  "properties": {
    "inputText": {
      "type": "string"
    }
  },
  "required": [
    "inputText"
  ]
}

Jurassic2-Ultra with id ai21.j2-ultra-v1 Jurassic-2 Ultra is AI21’s most powerful model for complex tasks that require advanced text generation and comprehension.

Json schema for request

{
  "$schema": "http://json-schema.org/draft-04/schema#",
  "type": "object",
  "properties": {
    "prompt": {
      "type": "string"
    },
    "maxTokens": {
      "type": "integer"
    },
    "temperature": {
      "type": "integer"
    },
    "topP": {
      "type": "integer"
    },
    "stopSequences": {
      "type": "array",
      "items": [
        {
          "type": "string"
        }
      ]
    },
    "presencePenalty": {
      "type": "object",
      "properties": {
        "scale": {
          "type": "integer"
        }
      },
      "required": [
        "scale"
      ]
    },
    "frequencyPenalty": {
      "type": "object",
      "properties": {
        "scale": {
          "type": "integer"
        }
      },
      "required": [
        "scale"
      ]
    }
  },
  "required": [
    "prompt",
    "maxTokens",
    "temperature",
    "topP",
    "stopSequences",
    "presencePenalty",
    "frequencyPenalty"
  ]
}

Jurassic2-Mid with id ai21.j2-mid-v1 Jurassic-2 Mid is less powerful than Ultra, yet carefully designed to strike the right balance between exceptional quality and affordability.

Json schema for request

{
  "$schema": "http://json-schema.org/draft-04/schema#",
  "type": "object",
  "properties": {
    "prompt": {
      "type": "string"
    },
    "maxTokens": {
      "type": "integer"
    },
    "temperature": {
      "type": "integer"
    },
    "topP": {
      "type": "integer"
    },
    "stopSequences": {
      "type": "array",
      "items": [
        {
          "type": "string"
        }
      ]
    },
    "presencePenalty": {
      "type": "object",
      "properties": {
        "scale": {
          "type": "integer"
        }
      },
      "required": [
        "scale"
      ]
    },
    "frequencyPenalty": {
      "type": "object",
      "properties": {
        "scale": {
          "type": "integer"
        }
      },
      "required": [
        "scale"
      ]
    }
  },
  "required": [
    "prompt",
    "maxTokens",
    "temperature",
    "topP",
    "stopSequences",
    "presencePenalty",
    "frequencyPenalty"
  ]
}

Claude Instant V1.2 with id anthropic.claude-instant-v1 A fast, affordable yet still very capable model, which can handle a range of tasks including casual dialogue, text analysis, summarization, and document question-answering.

Json schema for request

{
  "$schema": "http://json-schema.org/draft-04/schema#",
  "type": "object",
  "properties": {
    "prompt": {
      "type": "string"
    },
    "max_tokens_to_sample": {
      "type": "integer"
    },
    "stop_sequences": {
      "type": "array",
      "items": [
        {
          "type": "string"
        }
      ]
    },
    "temperature": {
      "type": "number"
    },
    "top_p": {
      "type": "integer"
    },
    "top_k": {
      "type": "integer"
    },
    "anthropic_version": {
      "type": "string"
    }
  },
  "required": [
    "prompt",
    "max_tokens_to_sample",
    "stop_sequences",
    "temperature",
    "top_p",
    "top_k",
    "anthropic_version"
  ]
}

Claude 2 with id anthropic.claude-v2 Anthropic’s highly capable model across a wide range of tasks from sophisticated dialogue and creative content generation to detailed instruction following.

Json schema for request

{
  "$schema": "http://json-schema.org/draft-04/schema#",
  "type": "object",
  "properties": {
    "prompt": {
      "type": "string"
    },
    "max_tokens_to_sample": {
      "type": "integer"
    },
    "stop_sequences": {
      "type": "array",
      "items": [
        {
          "type": "string"
        }
      ]
    },
    "temperature": {
      "type": "number"
    },
    "top_p": {
      "type": "integer"
    },
    "top_k": {
      "type": "integer"
    },
    "anthropic_version": {
      "type": "string"
    }
  },
  "required": [
    "prompt",
    "max_tokens_to_sample",
    "stop_sequences",
    "temperature",
    "top_p",
    "top_k",
    "anthropic_version"
  ]
}

Claude 2.1 with id anthropic.claude-v2:1 An update to Claude 2 that features double the context window, plus improvements across reliability, hallucination rates, and evidence-based accuracy in long document and RAG contexts.

Json schema for request

{
  "$schema": "http://json-schema.org/draft-04/schema#",
  "type": "object",
  "properties": {
    "prompt": {
      "type": "string"
    },
    "max_tokens_to_sample": {
      "type": "integer"
    },
    "stop_sequences": {
      "type": "array",
      "items": [
        {
          "type": "string"
        }
      ]
    },
    "temperature": {
      "type": "number"
    },
    "top_p": {
      "type": "integer"
    },
    "top_k": {
      "type": "integer"
    },
    "anthropic_version": {
      "type": "string"
    }
  },
  "required": [
    "prompt",
    "max_tokens_to_sample",
    "stop_sequences",
    "temperature",
    "top_p",
    "top_k",
    "anthropic_version"
  ]
}

Claude 3 Sonnet with id anthropic.claude-3-sonnet-20240229-v1:0 Claude 3 Sonnet by Anthropic strikes the ideal balance between intelligence and speed—particularly for enterprise workloads.

Json schema for request

{
  "$schema": "http://json-schema.org/draft-04/schema#",
  "type": "object",
  "properties": {
    "messages": {
      "type": "array",
      "items": [
        {
          "type": "object",
          "properties": {
            "role": {
              "type": "string"
            },
            "content": {
              "type": "array",
              "items": [
                {
                  "type": "object",
                  "properties": {
                    "type": {
                      "type": "string"
                    },
                    "text": {
                      "type": "string"
                    }
                  },
                  "required": [
                    "type",
                    "text"
                  ]
                }
              ]
            }
          },
          "required": [
            "role",
            "content"
          ]
        }
      ]
    },
    "max_tokens": {
      "type": "integer"
    },
    "anthropic_version": {
      "type": "string"
    }
  },
  "required": [
    "messages",
    "max_tokens",
    "anthropic_version"
  ]
}

Claude 3 Haiku with id anthropic.claude-3-haiku-20240307-v1:0 Claude 3 Haiku is Anthropic’s fastest, most compact model for near-instant responsiveness. It answers simple queries and requests with speed.

Json schema for request

{
  "$schema": "http://json-schema.org/draft-04/schema#",
  "type": "object",
  "properties": {
    "messages": {
      "type": "array",
      "items": [
        {
          "type": "object",
          "properties": {
            "role": {
              "type": "string"
            },
            "content": {
              "type": "array",
              "items": [
                {
                  "type": "object",
                  "properties": {
                    "type": {
                      "type": "string"
                    },
                    "text": {
                      "type": "string"
                    }
                  },
                  "required": [
                    "type",
                    "text"
                  ]
                }
              ]
            }
          },
          "required": [
            "role",
            "content"
          ]
        }
      ]
    },
    "max_tokens": {
      "type": "integer"
    },
    "anthropic_version": {
      "type": "string"
    }
  },
  "required": [
    "messages",
    "max_tokens",
    "anthropic_version"
  ]
}