Skip to content

Conversation

@arturantonnikau
Copy link
Contributor

No description provided.

# frozen_string_literal: true

module Clients
class NekkoaiClient
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You don't need custom client for Nekkoai - it's fully OpenAI compatible. You can use OpenAI client and just change config (URL).

DEFAULT_CONFIGURATION = {
'openai' => '{"model":"gpt-3.5-turbo","temperature":0.5}',
'ollama' => '{"model":"llama3.1", "stream": false}',
'nekkoai' => '{"model":"llama","max_tokens":200,"top_p":0.3,"stop":["4.","sushi"]}',
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Everything but "model": "llama" has only meaning in the context of the example it was used in. We definitely don't care about sushi in llamagator, or do we? :)

const DEFAULT_CONFIGURATION = {
openai: '{"model":"gpt-3.5-turbo","temperature":0.5}',
ollama: '{"model":"llama3.1", "stream": false}',
nekkoai: '{"model":"llama","max_tokens":200,"top_p":0.3,"stop":["4.","sushi"]}',
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Same: please leave just model specifier.

DEFAULT_CONFIGURATION = {
'openai' => '{"model":"gpt-3.5-turbo","temperature":0.5}',
'ollama' => '{"model":"llama3.1", "stream": false}',
'nekkoai' => '{"model":"llama","max_tokens":200,"top_p":0.3,"stop":["4.","sushi"]}',
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm not sure at this moment if nekkoai is correct name for this connector.
The company is named "Ainekko". The llama.cpp wrapper - NekkoAPI (or nekko-api).
@Sergyenko please advise.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants