Better avante config

This commit is contained in:
Marcelo Jacobus 2025-06-29 22:33:00 -03:00
parent d0940f2b4d
commit 43f6704416

View file

@ -18,20 +18,20 @@ return {
claude = {
endpoint = "https://api.anthropic.com",
model = "claude-3-5-sonnet-20241022",
timeout = 30000, -- Timeout in milliseconds
extra_request_body = {
temperature = 0,
max_tokens = 4096,
temperature = 0.3, -- Slightly increased for better balance
max_tokens = 2048, -- Reduced to save costs while still being effective
}
},
openai = {
endpoint = "https://api.openai.com/v1",
model = "gpt-4o", -- your desired model (or use gpt-4o, etc.)
timeout = 30000, -- Timeout in milliseconds, increase this for reasoning models
timeout = 30000,
extra_request_body = {
temperature = 0,
max_completion_tokens = 8192, -- Increase this to include reasoning tokens (for reasoning models)
temperature = 0.3,
max_completion_tokens = 2048, -- Reduced to save costs
}
--reasoning_effort = "medium", -- low|medium|high, only used for reasoning models
},
}
},