| Age | Commit message (Collapse) | Author |
|
Set the main ollama model to minimax-m2.1:cloud and the in-editor completion
model to qwen3-coder:480b-cloud (temperature 0.0 for deterministic completion),
and drop the commented-out alternative-model lists.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
|
|
|
|
|
|
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Manage ~/.config/hexai/config.toml from the dotfiles repo, mirroring how
helix, lazygit, and opencode are handled. The home_hexai task is
gated to Linux because hexai is currently only used on the Linux
workstation.
The deployed config sets Ollama Cloud (kimi-k2.6 at https://ollama.com)
as the default provider and uses gemma4:31b for in-code completion.
The API key is read from $HEXAI_OLLAMA_API_KEY (or $OLLAMA_API_KEY).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
|