summaryrefslogtreecommitdiffstats
path: root/models.yaml
diff options
context:
space:
mode:
authorsigoden <sigoden@gmail.com>2024-06-05 09:02:23 +0800
committerGitHub <noreply@github.com>2024-06-05 09:02:23 +0800
commit1ec6abfaee2fdc189b348b7e3a8145bd9a84da74 (patch)
tree6f68a860a39b0fbe87784de925b5a31e00e74e33 /models.yaml
parent71f2e94579511d7524f5534377001ab3f02a9597 (diff)
downloadaichat-1ec6abfaee2fdc189b348b7e3a8145bd9a84da74.tar.gz
feat: support RAG (#560)
* feat: support RAG * support more embeddings models and implement concurrent embedding api * show the progress of addings paths * ignore embedding context when saving message * embedding model max_chunk_size => default_chunk_size * support pdf and pandoc formats (docx, epub, ipynb)
Diffstat (limited to 'models.yaml')
-rw-r--r--models.yaml43
1 files changed, 43 insertions, 0 deletions
diff --git a/models.yaml b/models.yaml
index c306f1b..f4c980b 100644
--- a/models.yaml
+++ b/models.yaml
@@ -65,6 +65,16 @@
max_output_tokens: 4096
input_price: 60
output_price: 120
+ - name: text-embedding-3-large
+ mode: embedding
+ max_input_tokens: 8191
+ default_chunk_size: 8000
+ max_concurrent_chunks: 100
+ - name: text-embedding-3-small
+ mode: embedding
+ max_input_tokens: 8191
+ default_chunk_size: 8000
+ max_concurrent_chunks: 100
- platform: gemini
# docs:
@@ -100,6 +110,10 @@
output_price: 10.5
supports_vision: true
supports_function_calling: true
+ - name: text-embedding-004
+ mode: embedding
+ max_input_tokens: 2048
+ default_chunk_size: 2000
- platform: claude
# docs:
@@ -162,6 +176,10 @@
input_price: 8
output_price: 24
supports_function_calling: true
+ - name: mistral-embed
+ mode: embedding
+ max_input_tokens: 8092
+ default_chunk_size: 8000
- platform: cohere
# docs:
@@ -183,6 +201,16 @@
input_price: 3
output_price: 15
supports_function_calling: true
+ - name: embed-english-v3.0
+ mode: embedding
+ max_input_tokens: 512
+ default_chunk_size: 1000
+ max_concurrent_chunks: 96
+ - name: embed-multilingual-v3.0
+ mode: embedding
+ max_input_tokens: 512
+ default_chunk_size: 1000
+ max_concurrent_chunks: 96
- platform: perplexity
# docs:
@@ -281,6 +309,16 @@
input_price: 1.25
output_price: 3.75
supports_vision: true
+ - name: text-embedding-004
+ mode: embedding
+ max_input_tokens: 3072
+ default_chunk_size: 3000
+ max_concurrent_chunks: 5
+ - name: text-multilingual-embedding-002
+ mode: embedding
+ max_input_tokens: 3072
+ default_chunk_size: 3000
+ max_concurrent_chunks: 5
- platform: vertexai-claude
# docs:
@@ -509,6 +547,11 @@
input_price: 2.8
output_price: 2.8
supports_vision: true
+ - name: text-embedding-v2
+ mode: embedding
+ max_input_tokens: 2048
+ default_chunk_size: 2000
+ max_concurrent_chunks: 5
- platform: moonshot
# docs: