Faisal Amir 2a0601f75a
feat: remote engine management (#4364)
* feat: remote engine management

* chore: fix linter issue

* chore: remove unused imports

* fix: populate engines, models and legacy settings (#4403)

* fix: populate engines, models and legacy settings

* chore: legacy logics update configured remote engine

* fix: check exist path before reading

* fix: engines and models persist - race condition

* chore: update issue state

* test: update test cases

* chore: bring back Cortex extension settings

* chore: setup button gear / plus based apikey

* chore: fix remote engine from welcome screen

* chore: resolve linter issue

* chore: support request headers template

* chore: update engines using header_template instead of api_key_template

* chore: update models on changes

* fix: anthropic response template

* chore: fix welcome screen and debounce update value input

* chore: update engines list on changes

* chore: update engines list on change

* chore: update desc form add modal remote engines

* chore: bump cortex version to latest RC

* chore: fix linter

* fix: transform payload of Anthropic and OpenAI

* fix: typo

* fix: openrouter model id for auto routing

* chore: remove remote engine URL setting

* chore: add cohere engine and model support

* fix: should not clean on app launch - models list display issue

* fix: local engine check logic

* chore: bump app version to latest release 0.5.13

* test: fix failed tests

---------

Co-authored-by: Louis <louis@jan.ai>
2025-01-14 17:29:56 +07:00

29 lines
1.0 KiB
JSON

[
{
"model": "command-r-plus",
"object": "model",
"name": "Command R+",
"version": "1.0",
"description": "Command R+ is an instruction-following conversational model that performs language tasks at a higher quality, more reliably, and with a longer context than previous models. It is best suited for complex RAG workflows and multi-step tool use.",
"inference_params": {
"max_tokens": 4096,
"temperature": 0.7,
"stream": false
},
"engine": "cohere"
},
{
"model": "command-r",
"object": "model",
"name": "Command R",
"version": "1.0",
"description": "Command R is an instruction-following conversational model that performs language tasks at a higher quality, more reliably, and with a longer context than previous models. It can be used for complex workflows like code generation, retrieval augmented generation (RAG), tool use, and agents.",
"inference_params": {
"max_tokens": 4096,
"temperature": 0.7,
"stream": false
},
"engine": "cohere"
}
]