From ea9a6d8cba309d75e71f77af1be1b949b8bea8dc Mon Sep 17 00:00:00 2001
From: AAEE86 <33052466+AAEE86@users.noreply.github.com>
Date: Thu, 26 Sep 2024 19:08:59 +0800
Subject: [PATCH] feat: add stream for Gemini (#8678)

---
 .../google/llm/gemini-1.5-flash-8b-exp-0827.yaml         | 9 +++++++++
 .../google/llm/gemini-1.5-flash-exp-0827.yaml            | 9 +++++++++
 .../google/llm/gemini-1.5-flash-latest.yaml              | 9 +++++++++
 .../google/llm/gemini-1.5-pro-exp-0801.yaml              | 9 +++++++++
 .../google/llm/gemini-1.5-pro-exp-0827.yaml              | 9 +++++++++
 .../google/llm/gemini-1.5-pro-latest.yaml                | 9 +++++++++
 .../model_providers/google/llm/gemini-pro-vision.yaml    | 9 +++++++++
 .../model_providers/google/llm/gemini-pro.yaml           | 9 +++++++++
 8 files changed, 72 insertions(+)

diff --git a/api/core/model_runtime/model_providers/google/llm/gemini-1.5-flash-8b-exp-0827.yaml b/api/core/model_runtime/model_providers/google/llm/gemini-1.5-flash-8b-exp-0827.yaml
index bbc697e934e055..4e0209890a336a 100644
--- a/api/core/model_runtime/model_providers/google/llm/gemini-1.5-flash-8b-exp-0827.yaml
+++ b/api/core/model_runtime/model_providers/google/llm/gemini-1.5-flash-8b-exp-0827.yaml
@@ -32,6 +32,15 @@ parameter_rules:
     max: 8192
   - name: response_format
     use_template: response_format
+  - name: stream
+    label:
+      zh_Hans: 流式输出
+      en_US: Stream
+    type: boolean
+    help:
+      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
+      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
+    default: false
 pricing:
   input: '0.00'
   output: '0.00'
diff --git a/api/core/model_runtime/model_providers/google/llm/gemini-1.5-flash-exp-0827.yaml b/api/core/model_runtime/model_providers/google/llm/gemini-1.5-flash-exp-0827.yaml
index c5695e5dda8eb0..faabc5e4d13a73 100644
--- a/api/core/model_runtime/model_providers/google/llm/gemini-1.5-flash-exp-0827.yaml
+++ b/api/core/model_runtime/model_providers/google/llm/gemini-1.5-flash-exp-0827.yaml
@@ -32,6 +32,15 @@ parameter_rules:
     max: 8192
   - name: response_format
     use_template: response_format
+  - name: stream
+    label:
+      zh_Hans: 流式输出
+      en_US: Stream
+    type: boolean
+    help:
+      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
+      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
+    default: false
 pricing:
   input: '0.00'
   output: '0.00'
diff --git a/api/core/model_runtime/model_providers/google/llm/gemini-1.5-flash-latest.yaml b/api/core/model_runtime/model_providers/google/llm/gemini-1.5-flash-latest.yaml
index 24b1c5af8a3fd8..6a0344699a78b7 100644
--- a/api/core/model_runtime/model_providers/google/llm/gemini-1.5-flash-latest.yaml
+++ b/api/core/model_runtime/model_providers/google/llm/gemini-1.5-flash-latest.yaml
@@ -32,6 +32,15 @@ parameter_rules:
     max: 8192
   - name: response_format
     use_template: response_format
+  - name: stream
+    label:
+      zh_Hans: 流式输出
+      en_US: Stream
+    type: boolean
+    help:
+      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
+      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
+    default: false
 pricing:
   input: '0.00'
   output: '0.00'
diff --git a/api/core/model_runtime/model_providers/google/llm/gemini-1.5-pro-exp-0801.yaml b/api/core/model_runtime/model_providers/google/llm/gemini-1.5-pro-exp-0801.yaml
index 0a918e0d7b1ac3..97c68f7a18d91e 100644
--- a/api/core/model_runtime/model_providers/google/llm/gemini-1.5-pro-exp-0801.yaml
+++ b/api/core/model_runtime/model_providers/google/llm/gemini-1.5-pro-exp-0801.yaml
@@ -32,6 +32,15 @@ parameter_rules:
     max: 8192
   - name: response_format
     use_template: response_format
+  - name: stream
+    label:
+      zh_Hans: 流式输出
+      en_US: Stream
+    type: boolean
+    help:
+      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
+      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
+    default: false
 pricing:
   input: '0.00'
   output: '0.00'
diff --git a/api/core/model_runtime/model_providers/google/llm/gemini-1.5-pro-exp-0827.yaml b/api/core/model_runtime/model_providers/google/llm/gemini-1.5-pro-exp-0827.yaml
index 7452ce46e7dcb6..860e4816a163cc 100644
--- a/api/core/model_runtime/model_providers/google/llm/gemini-1.5-pro-exp-0827.yaml
+++ b/api/core/model_runtime/model_providers/google/llm/gemini-1.5-pro-exp-0827.yaml
@@ -32,6 +32,15 @@ parameter_rules:
     max: 8192
   - name: response_format
     use_template: response_format
+  - name: stream
+    label:
+      zh_Hans: 流式输出
+      en_US: Stream
+    type: boolean
+    help:
+      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
+      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
+    default: false
 pricing:
   input: '0.00'
   output: '0.00'
diff --git a/api/core/model_runtime/model_providers/google/llm/gemini-1.5-pro-latest.yaml b/api/core/model_runtime/model_providers/google/llm/gemini-1.5-pro-latest.yaml
index b3e1ecf3aff379..92cd6b310d07ed 100644
--- a/api/core/model_runtime/model_providers/google/llm/gemini-1.5-pro-latest.yaml
+++ b/api/core/model_runtime/model_providers/google/llm/gemini-1.5-pro-latest.yaml
@@ -32,6 +32,15 @@ parameter_rules:
     max: 8192
   - name: response_format
     use_template: response_format
+  - name: stream
+    label:
+      zh_Hans: 流式输出
+      en_US: Stream
+    type: boolean
+    help:
+      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
+      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
+    default: false
 pricing:
   input: '0.00'
   output: '0.00'
diff --git a/api/core/model_runtime/model_providers/google/llm/gemini-pro-vision.yaml b/api/core/model_runtime/model_providers/google/llm/gemini-pro-vision.yaml
index 075e484e469308..2d213d56adb9c7 100644
--- a/api/core/model_runtime/model_providers/google/llm/gemini-pro-vision.yaml
+++ b/api/core/model_runtime/model_providers/google/llm/gemini-pro-vision.yaml
@@ -27,6 +27,15 @@ parameter_rules:
     default: 4096
     min: 1
     max: 4096
+  - name: stream
+    label:
+      zh_Hans: 流式输出
+      en_US: Stream
+    type: boolean
+    help:
+      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
+      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
+    default: false
 pricing:
   input: '0.00'
   output: '0.00'
diff --git a/api/core/model_runtime/model_providers/google/llm/gemini-pro.yaml b/api/core/model_runtime/model_providers/google/llm/gemini-pro.yaml
index 4e9f59e7da94f0..e2f487c1ee9219 100644
--- a/api/core/model_runtime/model_providers/google/llm/gemini-pro.yaml
+++ b/api/core/model_runtime/model_providers/google/llm/gemini-pro.yaml
@@ -31,6 +31,15 @@ parameter_rules:
     max: 2048
   - name: response_format
     use_template: response_format
+  - name: stream
+    label:
+      zh_Hans: 流式输出
+      en_US: Stream
+    type: boolean
+    help:
+      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
+      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
+    default: false
 pricing:
   input: '0.00'
   output: '0.00'