Merge branch 'feat/model-runtime' into deploy/dev

langgenius · Jan 2, 2024 · 9c6950f · 9c6950f
2 parents 53837db + 71cb2e0
commit 9c6950f
Show file tree

Hide file tree

Showing 5 changed files with 6 additions and 6 deletions.
diff --git a/api/core/agent/agent/calc_token_mixin.py b/api/core/agent/agent/calc_token_mixin.py
@@ -29,7 +29,7 @@ def get_message_rest_tokens(self, model_config: ModelConfigEntity, messages: Lis
             if (parameter_rule.name == 'max_tokens'
                     or (parameter_rule.use_template and parameter_rule.use_template == 'max_tokens')):
                 max_tokens = (model_config.parameters.get(parameter_rule.name)
-                              or model_config.parameters.get(parameter_rule.use_template))
+                              or model_config.parameters.get(parameter_rule.use_template)) or 0
 
         if model_context_tokens is None:
             return 0

diff --git a/api/core/agent/agent/openai_function_call.py b/api/core/agent/agent/openai_function_call.py
@@ -79,7 +79,7 @@ def should_use_agent(self, query: str):
             if (parameter_rule.name == 'max_tokens'
                     or (parameter_rule.use_template and parameter_rule.use_template == 'max_tokens')):
                 original_max_tokens = (self.model_config.parameters.get(parameter_rule.name)
-                              or self.model_config.parameters.get(parameter_rule.use_template))
+                              or self.model_config.parameters.get(parameter_rule.use_template)) or 0
 
         self.model_config.parameters['max_tokens'] = 40
 

diff --git a/api/core/app_runner/app_runner.py b/api/core/app_runner/app_runner.py
@@ -41,7 +41,7 @@ def get_pre_calculate_rest_tokens(self, app_record: App,
             if (parameter_rule.name == 'max_tokens'
                     or (parameter_rule.use_template and parameter_rule.use_template == 'max_tokens')):
                 max_tokens = (model_config.parameters.get(parameter_rule.name)
-                              or model_config.parameters.get(parameter_rule.use_template))
+                              or model_config.parameters.get(parameter_rule.use_template)) or 0
 
         if model_context_tokens is None:
             return -1
@@ -85,7 +85,7 @@ def recale_llm_max_tokens(self, model_config: ModelConfigEntity,
             if (parameter_rule.name == 'max_tokens'
                     or (parameter_rule.use_template and parameter_rule.use_template == 'max_tokens')):
                 max_tokens = (model_config.parameters.get(parameter_rule.name)
-                              or model_config.parameters.get(parameter_rule.use_template))
+                              or model_config.parameters.get(parameter_rule.use_template)) or 0
 
         if model_context_tokens is None:
             return -1

diff --git a/api/core/prompt/prompt_transform.py b/api/core/prompt/prompt_transform.py
@@ -404,7 +404,7 @@ def _calculate_rest_token(self, prompt_messages: list[PromptMessage], model_conf
                 if (parameter_rule.name == 'max_tokens'
                         or (parameter_rule.use_template and parameter_rule.use_template == 'max_tokens')):
                     max_tokens = (model_config.parameters.get(parameter_rule.name)
-                                  or model_config.parameters.get(parameter_rule.use_template))
+                                  or model_config.parameters.get(parameter_rule.use_template)) or 0
 
             rest_tokens = model_context_tokens - max_tokens - curr_message_tokens
             rest_tokens = max(rest_tokens, 0)

diff --git a/api/services/app_model_config_service.py b/api/services/app_model_config_service.py
@@ -163,7 +163,7 @@ def validate_configuration(cls, tenant_id: str, account: Account, config: dict,
 
         # model.mode
         if model_mode:
-            config['model']["mode"] = model_mode
+            config['model']["mode"] = model_mode.value
         else:
             config['model']["mode"] = "completion"