blob: c6ebb1998a370f77fb50d1afee6f5c2b618c1692 (
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
|
[app]
# LLM used for internal operations, like deriving conversation names
fast_llm = "azure/gpt-4o-mini"
# LLM used for user-facing output, like RAG replies
quality_llm = "azure/gpt-4o"
# LLM used for ingesting visual inputs
vlm = "azure/gpt-4o"
# LLM used for transcription
audio_lm = "azure/whisper-1"
# Reasoning model, used for `research` agent
reasoning_llm = "azure/o3-mini"
# Planning model, used for `research` agent
planning_llm = "azure/o3-mini"
[embedding]
base_model = "azure/text-embedding-3-small"
[completion_embedding]
base_model = "azure/text-embedding-3-small"
[ingestion]
provider = "unstructured_local"
strategy = "auto"
chunking_strategy = "by_title"
new_after_n_chars = 2_048
max_characters = 4_096
combine_under_n_chars = 1_024
overlap = 1_024
document_summary_model = "azure/gpt-4o-mini"
automatic_extraction = true # enable automatic extraction of entities and relations
[ingestion.extra_parsers]
pdf = "zerox"
[ingestion.chunk_enrichment_settings]
generation_config = { model = "azure/gpt-4o-mini" }
[orchestration]
provider = "hatchet"
kg_creation_concurrency_limit = 32
ingestion_concurrency_limit = 4
kg_concurrency_limit = 8
|
