From b4a40e3fedb438570770a224b890ea24f6e660a9 Mon Sep 17 00:00:00 2001
From: sigoden <sigoden@gmail.com>
Date: Sat, 18 May 2024 19:06:21 +0800
Subject: feat: support function calling (#514)

* feat: support function calling

* fix on Windows OS

* implement multi-steps function calling

* fix on Windows OS

* add error for client not support function calling

* refactor message data structure and make claude client supporting function calling

* support reuse previous call results

* improve error handling for function calling

* use prefix `may_`  as indicator for `execute` type fucntions
---
 models.yaml | 25 +++++++++++++++++++++++--
 1 file changed, 23 insertions(+), 2 deletions(-)

(limited to 'models.yaml')

diff --git a/models.yaml b/models.yaml
index 903e285..3c78470 100644
--- a/models.yaml
+++ b/models.yaml
@@ -15,33 +15,39 @@
       max_output_tokens: 4096
       input_price: 0.5
       output_price: 1.5
+      supports_function_calling: true
     - name: gpt-3.5-turbo-1106
       max_input_tokens: 16385
       max_output_tokens: 4096
       input_price: 1
       output_price: 2
+      supports_function_calling: true
     - name: gpt-4o
       max_input_tokens: 128000
       max_output_tokens: 4096
       input_price: 5
       output_price: 15
       supports_vision: true
+      supports_function_calling: true
     - name: gpt-4-turbo
       max_input_tokens: 128000
       max_output_tokens: 4096
       input_price: 10
       output_price: 30
       supports_vision: true
+      supports_function_calling: true
     - name: gpt-4-turbo-preview
       max_input_tokens: 128000
       max_output_tokens: 4096
       input_price: 10
       output_price: 30
+      supports_function_calling: true
     - name: gpt-4-1106-preview
       max_input_tokens: 128000
       max_output_tokens: 4096
       input_price: 10
       output_price: 30
+      supports_function_calling: true
     - name: gpt-4-vision-preview
       max_input_tokens: 128000
       max_output_tokens: 4096
@@ -73,6 +79,7 @@
       max_output_tokens: 2048
       input_price: 0.5
       output_price: 1.5
+      supports_function_calling: true
     - name: gemini-1.0-pro-vision-latest
       max_input_tokens: 12288
       max_output_tokens: 4096
@@ -85,12 +92,14 @@
       input_price: 0.35
       output_price: 0.53
       supports_vision: true
+      supports_function_calling: true
     - name: gemini-1.5-pro-latest
       max_input_tokens: 1048576
       max_output_tokens: 8192
       input_price: 3.5
       output_price: 10.5
       supports_vision: true
+      supports_function_calling: true
 
 - platform: claude
   # docs:
@@ -106,6 +115,7 @@
       input_price: 15
       output_price: 75
       supports_vision: true
+      supports_function_calling: true
     - name: claude-3-sonnet-20240229
       max_input_tokens: 200000
       max_output_tokens: 4096
@@ -113,6 +123,7 @@
       input_price: 3
       output_price: 15
       supports_vision: true
+      supports_function_calling: true
     - name: claude-3-haiku-20240307
       max_input_tokens: 200000
       max_output_tokens: 4096
@@ -120,6 +131,7 @@
       input_price: 0.25
       output_price: 1.25
       supports_vision: true
+      supports_function_calling: true
 
 - platform: mistral
   # docs:
@@ -149,6 +161,7 @@
       max_input_tokens: 32000
       input_price: 8
       output_price: 24
+      supports_function_calling: true
 
 - platform: cohere
   # docs:
@@ -163,11 +176,13 @@
       max_output_tokens: 4000
       input_price: 0.5
       output_price: 1.5
+      supports_function_calling: true
     - name: command-r-plus
       max_input_tokens: 128000
       max_output_tokens: 4000
       input_price: 3
       output_price: 15
+      supports_function_calling: true
 
 - platform: perplexity
   # docs:
@@ -242,12 +257,13 @@
   # notes:
   #   - get max_output_tokens info from models doc
   models:
-    - name: gemini-1.0-pro
+    - name: gemini-1.0-pro-002
       max_input_tokens: 24568
       max_output_tokens: 8192
       input_price: 0.125
       output_price: 0.375
-    - name: gemini-1.0-pro-vision
+      supports_function_calling: true
+    - name: gemini-1.0-pro-vision-001
       max_input_tokens: 14336
       max_output_tokens: 2048
       input_price: 0.125
@@ -387,6 +403,7 @@
   # docs:
   #   - https://replicate.com/docs
   #   - https://replicate.com/pricing
+  #   - https://replicate.com/docs/reference/http
   # notes:
   #   - max_output_tokens is required but unknown
   models:
@@ -695,20 +712,24 @@
       max_input_tokens: 16385
       input_price: 0.5
       output_price: 1.5
+      supports_function_calling: true
     - name: openai/gpt-4o
       max_input_tokens: 128000
       input_price: 5
       output_price: 15
       supports_vision: true
+      supports_function_calling: true
     - name: openai/gpt-4-turbo
       max_input_tokens: 128000
       input_price: 10
       output_price: 30
       supports_vision: true
+      supports_function_calling: true
     - name: openai/gpt-4-turbo-preview
       max_input_tokens: 128000
       input_price: 10
       output_price: 30
+      supports_function_calling: true
     - name: openai/gpt-4-vision-preview
       max_input_tokens: 128000
       max_output_tokens: 4096
-- 
cgit v1.2.3