blob: 5226eebf8b8a7713bd552367712e0e2a0300e404 (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
|
[app]
# LLM used for internal operations, like deriving conversation names
fast_llm = "openai/llama3.1" ### NOTE - RECOMMENDED TO USE `openai` with `api_base = "http://localhost:11434/v1"` for best results, otherwise `ollama` with `litellm` is acceptable
# LLM used for user-facing output, like RAG replies
quality_llm = "openai/llama3.1"
# LLM used for ingesting visual inputs
vlm = "openai/llama3.1" # TODO - Replace with viable candidate
# LLM used for transcription
audio_lm = "openai/llama3.1" # TODO - Replace with viable candidate
# Reasoning model, used for `research` agent
reasoning_llm = "openai/llama3.1"
# Planning model, used for `research` agent
planning_llm = "openai/llama3.1"
[embedding]
provider = "ollama"
base_model = "mxbai-embed-large"
base_dimension = 1_024
batch_size = 128
add_title_as_prefix = true
concurrent_request_limit = 2
[completion_embedding]
provider = "ollama"
base_model = "mxbai-embed-large"
base_dimension = 1_024
batch_size = 128
add_title_as_prefix = true
concurrent_request_limit = 2
[agent]
tools = ["search_file_knowledge"]
[completion]
provider = "litellm"
concurrent_request_limit = 1
[completion.generation_config]
temperature = 0.1
top_p = 1
max_tokens_to_sample = 1_024
stream = false
api_base = "http://localhost:11434/v1"
|
