Drezil/snippets
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

tweaked avante, removed example, added generate_overview.bash

Browse Source
This commit is contained in:
Nicole Dresselhaus
2025-07-01 20:34:13 +02:00
parent ace703ea1d
commit 41e1067647
5 changed files with 81 additions and 227 deletions
Download Patch File Download Diff File Expand all files Collapse all files

40
nvim-config/lua/plugins/avante.lua
View File

@ -9,26 +9,18 @@ return {
behaviour = {
enable_cursor_planning_mode = true, -- enable cursor planning mode!
},
ollama = {
endpoint = "http://gpu.dighist.geschichte.hu-berlin.de:11434",
model = "cogito:14b", -- your desired model (or use gpt-4o, etc.)
timeout = 30000, -- Timeout in milliseconds, increase this for reasoning models
temperature = 0,
max_completion_tokens = 8192, -- Increase this to include reasoning tokens (for reasoning models)
stream = true,
thinking = true,
system_prompt = "Enable deep thinking subroutine.",
-- reasoning_effort = "high", -- low|medium|high, only used for reasoning models
},
rag_service = {
enabled = true, -- Enables the RAG service
host_mount = os.getenv("HOME"), -- Host mount path for the rag service
provider = "ollama", -- The provider to use for RAG service (e.g. openai or ollama)
llm_model = "qwen3:32b", -- The LLM model to use for RAG service
embed_model = "nomic-embed-text", -- The embedding model to use for RAG service
endpoint = "http://gpu.dighist.geschichte.hu-berlin.de:11434", -- The API endpoint for RAG service
},
vendors = {
providers = {
ollama = {
endpoint = "http://gpu.dighist.geschichte.hu-berlin.de:11434",
model = "cogito:32b", -- your desired model (or use gpt-4o, etc.)
timeout = 30000, -- Timeout in milliseconds, increase this for reasoning models
temperature = 0,
max_completion_tokens = 40000, -- Increase this to include reasoning tokens (for reasoning models)
stream = true,
thinking = true,
--system_prompt = "Enable deep thinking subroutine.",
-- reasoning_effort = "high", -- low|medium|high, only used for reasoning models
},
deepthink = {
__inherited_from = "ollama",
model = "qwen3:32b",
@ -36,6 +28,14 @@ return {
reasoning_effort = "high",
},
},
rag_service = {
enabled = true, -- Enables the RAG service
host_mount = os.getenv("HOME"), -- Host mount path for the rag service
provider = "ollama", -- The provider to use for RAG service (e.g. openai or ollama)
llm_model = "cogito", -- The LLM model to use for RAG service
embed_model = "nomic-embed-text", -- The embedding model to use for RAG service
endpoint = "http://gpu.dighist.geschichte.hu-berlin.de:11434", -- The API endpoint for RAG service
},
},
-- if you want to build from source then do `make BUILD_FROM_SOURCE=true`
build = "make",