LLM providerkdeps asset
kdeps-providers-cohere
v1.0.0kdeps provider: cohere
set providers / shelf model
PurposeCohere
LLM context: none - kdeps uses it without sending it to the model
Where it runs
Part of the providers set, used in workflow and agent mode. OpenAI-compatible LLM providers: base URL, chat path, API key setting and default model.
- a chat call names a backend
- the provider gives the URL, chat path and keyproviders/cohere
- kdeps sends the request
- the reply comes back
Loaded by: kdeps (agent loop), kdeps run, kdeps validate, kdeps doctor, kdeps edit, kdeps env, kdeps exec, kdeps test
In the code: kdeps (agent loop)llm.AutoRouterPickconfig.CloudLLMProvidersconfig.assetProviders
Install
A kdeps asset: installed with the first-run profile that includes its set, or the first time a command needs it, and updated without a new release.
kdeps install providers/cohere
Pin a version with kdeps install providers/cohere@<version>.
Contents
- name
- cohere
- label
- Cohere
- order
- 75
- configKey
- cohere_api_key
- envVar
- COHERE_API_KEY
- defaultModel
- command-r-plus
- url
- https://api.cohere.com
- chatPath
- /compatibility/v1/chat/completions
- embeddings
- false
hosts:
- cohere.com
# OpenAI-compatible LLM provider: kdeps registers a chat backend named after
# this file and reads its key from llm.cohere_api_key in ~/.kdeps/config.yaml.
# Cohere serves its chat models at this OpenAI-compatible path.
version: 1.0.0
name: cohere
label: Cohere
order: 75 # position in provider lists and the setup prompt
configKey: cohere_api_key
envVar: COHERE_API_KEY
defaultModel: command-r-plus
url: https://api.cohere.com
chatPath: /compatibility/v1/chat/completions
embeddings: false # embeddings use Cohere's native /embed API, not this path
hosts: [cohere.com]
Versions
| Version | Published | Status |
|---|---|---|
| 1.0.0 | 10/22/2026 | active |
Details
- Author
- kdeps
- License
- Apache-2.0
- Latest Version
- 1.0.0
- Published
- 10/22/2026
Tags
kdepsprovider