@gapi/openai

Installation

npm install @gapi/openai

Usage

import { Module } from '@gapi/core';
import { OpenAIModue } from '@gapi/openai';

@Module({
  imports: [
    OpenAIModue.forRoot({
      organization: 'org-ahnoIvsiO25Jq4hXeMxO3Fo8',
      apiKey: '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',
    })
  ]
})
export class AppModule {}

Exposed graphql mutation after importing

mutation {
  createCompletion(
    payload: {
      model: "text-davinci-003"
      prompt: "Can you write a description for a trello like card based on this text 'Create home landing page'"
      max_tokens: 2048
    }
  ) {
    id
    object
    created
    model
    choices {
      text
      index
      logprobs
      finish_reason
    }
    usage {
      prompt_tokens
      completion_tokens
      total_tokens
    }
  }
}

Create Chat Completion prompt

mutation {
  createChatCompletion(
    payload: {
      model: "gpt-3.5-turbo"
      prompt: "Can you write a description for card based on this text 'Create inside user portal page which will show when user log in to the system'"
      max_tokens: 2048
    }
  ) {
    id
    object
    created
    model
    choices {
      text
      message {
        role
        content
      }
      index
      logprobs
      finish_reason
    }
    usage {
      prompt_tokens
      completion_tokens
      total_tokens
    }
  }
}

Extending the OpenAPI graphql endpoints using internal provided injection token OpenAI

import { Controller, GraphQLNonNull, Inject, Mutation, Type } from '@gapi/core';
import {
  CreateCompletionInputType,
  CreateCompletionRequest,
  CreateCompletionType,
  OpenAI,
} from '@gapi/openai';
import { from } from 'rxjs';
import { map } from 'rxjs/operators';

@Controller()
export class CustomControllerController {
  constructor(@Inject(OpenAI) private openai: OpenAI) {}

  @Type(CreateCompletionType)
  @Mutation({
    payload: {
      type: new GraphQLNonNull(CreateCompletionInputType),
    },
  })
  createCompletion(root, { payload }: { payload: CreateCompletionRequest }) {
    return from(
      this.openai.createCompletion({
        ...payload,
        max_tokens: payload.max_tokens ?? 2048,
      }),
    ).pipe(map((res) => res.data));
  }


  @Type(CreateCompletionType)
  @Mutation({
    payload: {
      type: new GraphQLNonNull(CreateCompletionInputType),
    },
  })
  createChatCompletion(
    root,
    { payload }: { payload: CreateCompletionRequest },
  ) {
    return from(
      this.openai.createChatCompletion({
        messages: [{
          role: 'user',
          content: payload.prompt as string
        }],
        max_tokens: payload.max_tokens ?? 2048,
      }),
    ).pipe(map((res) => res.data));
  }
}