Multiple and local LLM support

[karthink]

EDIT: These instructions are obsolete as of v0.5.0, which adds support for GPT4All, Ollama and Azure. Please refer to the README instead.

---

I've added support for gptel-backends, which should make gptel work with most LLMs that offer REST APIs, including Azure and local LLMs like PrivateGPT and GPT4All. The design is still fluid and the integration is incomplete (no transient menu support yet), but I'd appreciate if someone with access to other LLMs can test it.

How to use it:

• Clone the multi-llm branch of gptel.

• Define a gptel-backendlike this:

  (defvar gptel--azure
  (gptel-make-backend
  "Azure"
  :host "api.openai.com"
  :endpoint "/openai/deployments/deploymentname/chat/completions?api-version=2023-03-15-preview"
  :header (lambda () '(("api-key" . ,(gptel--api-key))))
  :key #'gptel--api-key
  :protocol "https"
  :models '("gpt-3.5-turbo" "gpt-3.5-turbo-16k" "gpt-4" "gpt-4-32k")))

  For GPT4All:

  (defvar gptel--gpt4all
  (gptel-make-backend
  "GPT4All"
  :host "localhost:4891"
  :endpoint "/v1/chat/completions"
  :protocol "http"
  :models '("gpt4all-j-v1.3-groovy")))

  There are more examples in gptel-backends.el.

• Set this as the default backend, or set it locally in a gptel buffer:

  (setq-default gptel-backend gptel--azure) ; or whatever you defined

  OR in any chat buffer:

  (setq gptel-backend gptel--azure) ; or whatever you defined

• Use gptel as usual.

[karthink]

Update: Added multi-llm integration with the transient interface.