GenerativeĀ AI has leapt from research papers to daily business realityā and SAP is surfing that wave at full speed.Ā InĀ this handsāon series, Iāll show you how to spin up a custom AI agent on SAPĀ AIĀ Core in minutes, then grow it into a productionāready assetāwithout drowning in theory.
ēę AI ćÆē 究室ć®å¤ćøé£ć³åŗććä»ćććøćć¹ē¾å “ć®åøøčćå”ćęæćć¦ćć¾ććSAPćÆå Øéåć§ćć®ę³¢ć«ä¹ć£ć¦ćć¾ćććć®ććć°ć·ćŖć¼ćŗć§ćÆćSAPĀ AIĀ Core ć®ę¢å®ć¢ćć«ćęéć§å¼ć³åŗććå®åć§ä½æćć AI ćØć¼ćøć§ć³ććøę”å¼µććāē§éćć³ćŗćŖć³āććå±ććć¾ćć
Ā
šWhat Youāll Learn in This Series / ę¬ć·ćŖć¼ćŗć§å¦ć¹ćććØ
How to spin up a custom AIāÆagent on SAPāÆAIĀ Core in minutes
SAPĀ AIĀ Core äøć§ć«ć¹ćæć AI ćØć¼ćøć§ć³ćć āē§éā ć§åććę¹ę³
Handsāon with LangChain, Google Search Tool, RAG, and Streamlit
LangChainć»Google ę¤ē“¢ćć¼ć«ć»RAGć»Streamlit ć使ć£ćå®č£
Exposing the agent as a RESTĀ API and rebuilding the UI in SAPUI5/Fiori
ćØć¼ćøć§ć³ćć RESTĀ API åććUI ć SAPUI5/Fiori ć«č¼ćęæććęé
Time Commitment / å¦ēæęé
Each part is designed to be completed in 10ā15Ā minutes.
åē« ćÆ 10ā15 å ć§čŖććļ¼ęćåććććäŗå®ćć¦ćć¾ćć
Ā
šŗļø Series Roadmap / é£č¼ćć¼ćććć
Part 0 Prologue / ćććć¼ć°Ā Part 1 Env Setup: SAP AICore & AI Launchpad / ē°å¢ę§ēÆ: SAP AI CorećØAI LaunchpadĀ [current blog / ē¾åØć®ććć°]Part 2 Building a Chat Model withĀ LangChain / LangChain ć§ćć£ććć¢ćć«ćę§ēÆPart 3 AgentĀ Tools: Integrating GoogleĀ Search / Agentćć¼ć«: GoogleĀ ę¤ē“¢ćēµ±åPart 4 RAG Basics ā : File Upload & HANA Cloud VectorEngine / RAG åŗē¤ ā : ćć”ć¤ć«ć¢ćććć¼ć & HANA Cloud ććÆćæć¼ćØć³ćøć³Part 5 RAG Basics ā”: Retriever Tool / RAG åŗē¤ ā”: Retriever Toolć®ę§ēÆPart 6 Streamlit UI Prototype / Streamlit ć§ćć£ććUIPart 7 Expose as a RESTĀ API / AI Agent ć RESTĀ API åPart 8 Rebuild theĀ UI withĀ SAPUI5 / SAPUI5 ć§ćć£ććUI ćåę§ēÆ
Note / 注čØ
Subsequent blogs will be published soon.
ē¶ē·ØćÆé ꬔå
¬éäŗå®ć§ćć
EnvāÆSetup: SAP AICore & AI Launchpad
ē°å¢ę§ēÆ: SAP AI CorećØAI Launchpad
1Ā | Overview / ę¦č¦
In this chapter, weāll connect to SAP AI Launchpad using the service key created for the SAP AI Core instance. Then, weāll proceed to deploy the LLM for running a chat model.
ę¬ē« ć§ćÆćSAP AI Core ć¤ć³ć¹ćæć³ć¹ć«åƾćć¦ä½ęćććµć¼ćć¹ćć¼ć使ććSAP AI Launchpad ć«ę„ē¶ćć¾ćććć®å¾ććć£ććēØ LLM ćåććććć®ćććć¤ć”ć³ććå®ę½ćć¾ćć
Ā
2 | Prerequisites / äŗåęŗå
BTP sub-account / BTP ćµćć¢ć«ć¦ć³ćSAP AI Core instance / SAP AI Core ć¤ć³ć¹ćæć³ć¹SAP AI LaunchPad Subscription / SAP AI Launchpadć®ćµćć¹ćÆćŖćć·ć§ć³Python 3.13 and pip / Python 3.13ē°å¢ & pipVSCode, BAS or any IDE / VSCodećBASćŖć©ć®IDE
Ā
3Ā | StepĀ 1Ā āĀ Create a ServiceĀ Key / ServiceĀ Key ć®ä½ę
Creating a serviceĀ key gives your local scripts a secure, OAuthābased passport into SAPĀ AIĀ Core.Ā Without it, Python SDK calls have no way to authenticate or discover endpoints.
ćµć¼ćć¹ćć¼ćÆććć¼ć«ć«ć®ć¹ćÆćŖććć SAPĀ AIĀ Core ć«å®å Øć㤠OAuth ćć¼ć¹ć§ć¢ćÆć»ć¹ććććć® āćć¹ćć¼ćā ć§ćććććē”ććØćPython SDK ćÆćØć³ććć¤ć³ććčŖčؼę å ±ćåå¾ć§ćć¾ććć
Open your AIāÆCore instance in BTP Cockpit ā Instances &āÆSubscriptions.
BTPāÆCockpit ć® Instances &āÆSubscriptions ć§ AIāÆCore ć¤ć³ć¹ćæć³ć¹ćéćć¾ćć
Go to the ServiceĀ Keys tab and click Create. Give it any name.
ServiceĀ Keys ćæćć§ Create ććÆćŖććÆććä»»ęć®ååćå
„åćć¦ä½ęćć¾ćć
Download the generated JSON; youāll need it later.
ēęććć JSON ććć¦ć³ćć¼ććć¦äæē®”ćć¾ćć
NOTE ā Resource Group
When calling SAPĀ AIĀ Core from Python you must set an extra environment variable named RESOURCE_GROUP in addition to the values contained in the service key. In this tutorial weāll simply point it to the builtāin default group.
Python ććå¼ć³åŗćéć«ćÆćµć¼ćć¹ćć¼ć«å«ć¾ćć¦ććå¤ę°ä»„å¤ć« Resource Group ćØå¼ć°ććē°å¢å¤ę°ćåæ
č¦ć§ćććę¬ćć„ć¼ććŖć¢ć«ć§ćÆę¢å®ć® defaultĀ ćęå®ćć¾ćć
4Ā | StepĀ 2Ā āĀ Connect AIāÆLaunchpad / AIāÆLaunchpad ćØć®ę„ē¶
You can deploy an LLM (Large Language Model) directly to SAP AI Core using the service key you created earlier. However, in this guide, we’ll use SAP AI Launchpad to make the deployment process easier.
To begin, you’ll need to establish a secure connection route from SAP AI Launchpad to SAP AI Core.
å ć»ć©ä½ęćć SAP AI Core ć®ćµć¼ćć¹ćć¼ć使ćć°ćSAP AI Core ć«ē“ę„ć¢ćÆć»ć¹ć㦠LLMććććć¤ććććØćåÆč½ć§ćććć ććććć§ćÆććē°”åćŖę¹ę³ćØćć¦ćSAP AI Launchpadćć使ć£ć¦ćććć¤ćå®ę½ćć¾ććć¾ććÆćSAP AI Launchpad ćć SAP AI Core ćøå®å Øć«ę„ē¶ććććć®ēµč·Æćęŗåćć¾ćććć
From Subscriptions open AIāÆLaunchpad ā GoĀ toĀ Application.
Subscriptions ćć AIāÆLaunchpad ā GoĀ toĀ Application ć§čµ·åćć¾ćć
Click Add (topāright) ā API Connection and upload the ServiceĀ Key JSON.
å³äøć® Add ā API Connection ćéøć³ćServiceĀ Key JSON ćć¢ćććć¼ććć¾ćć
Select the default Resource Group in the side panel and save.
ćµć¤ćććć«ć§ Resource Group ć« default ćéøęćć¦äæåćć¾ćć
Ā
5Ā | StepĀ 3Ā āĀ Configure &Ā Deploy an LLM / ć¢ćć«čØå®ćØćććć¤
On the “Configuration” screen of SAP Launchpad, you can select the type of foundation model to use. Think of this as creating a model profile that can be shared across multiple deployments.
Letās go ahead and create the configuration. In this case, weāll use the default foundation_models scenario and configure it to use gapt-4o-mini.
SAP Launchpad ć®ćčØå®ćē»é¢ć§ćÆć使ēØććåŗē¤ć¢ćć«ć®ēØ®é”ćŖć©ćéøć³ć¾ćććććÆćč¤ę°ć®ćććć¤ć”ć³ćć§å ±ęć§ćć āć¢ćć«ć®ćććć”ć¤ć«ā ć®ćććŖćć®ćØčćććØēč§£ććććć§ćććć
ććć§ćÆčØå®ćä½ęćć¦ćæć¾ććććä»åćÆćććć©ć«ćć§ęä¾ććć¦ćć foundation_models ćØććć·ććŖćŖćéøęććgapt-4o-mini ćęå®ćć¾ćć
Navigate MLāÆOperations ā Settings and click Create.
å·¦ć”ćć„ć¼ MLāÆéēØ ā čØå® ć§ ä½ę ććÆćŖććÆćć¾ćć
Fill in: NameĀ =Ā <anyānameāyouālike>, ScenarioĀ =Ā foundation_models, VersionĀ =Ā 0.0.1, ExecutableĀ =Ā azure-openai.
čØå®å <ćčŖē±ć«>, ć·ććŖćŖ foundation_models, ćć¼ćøć§ć³ 0.0.1, å®č”åÆč½ azure-openai ćå
„åćć¾ćć
On the parameter screen leave modelName = gpt-4o-mini, modelVersion = latest and click Next ā Create.
ćć©ć”ć¼ćæē»é¢ć§ćÆ modelName = gpt-4o-mini, modelVersion = latest ććć®ć¾ć¾ć« 欔㸠ā ä½ęć
Youāll land on the new Setting detail pageācheck that all fields look correct; no extra edits needed.
ä½ęå¾ć«čØå®č©³ē“°ć®ē»é¢ćøé·ē§»ćć¾ććå
„åå
容ć«čŖ¤ćććŖćććć£ćØē¢ŗčŖćć¾ćććļ¼åŗę¬ēć«ćć®ć¾ć¾ć§ OK ć§ćļ¼ć
ClickĀ CreateĀ DeploymentĀ (topāright), breeze through the wizard by clickingĀ NextĀ untilĀ Create.
å³äøć®Ā ććććć¤ć”ć³ćä½ęćĀ ćę¼ććć¦ć£ć¶ć¼ććÆē¹ć«å¤ę“ććĀ ćꬔćøćĀ āĀ ćä½ęćĀ ć§é²ćć¾ćć
Wait untilĀ Status =Ā Runningāhit the Refresh icon every few seconds. Copy theĀ DeploymentĀ ID shown in the title bar for later use.
ć¹ćć¼ćæć¹ćĀ ćå®č”äøćĀ ć«ćŖćć¾ć§ćę“ę°ććæć³ć§ę°ē§ććć«ćŖćć¬ćć·ć„ćć¾ćććæć¤ćć«ćć¼ć«č”Øē¤ŗćććĀ DeploymentĀ IDĀ ćę§ćć¦ććć¾ćććć
By the way, there are various LLMs available to run on SAP AI Core. You can browse them in the Generative AI Hub > Model Library section.
ć”ćŖćæć«ćSAP AI Core äøć§å©ēØåÆč½ćŖ LLM ćÆćć¾ćć¾ć§ćććēę AI ćć > ć¢ćć«ć©ć¤ćć©ćŖćē»é¢ćéććØććććć®äøč¦§ć確čŖć§ćć¾ćć
The Leaderboard also allows you to compare models based on accuracy, token cost, and other metrics.
ć¾ćććŖć¼ćć¼ćć¼ćę©č½ć§ćÆćåć¢ćć«ć®ē²¾åŗ¦ććć¼ćÆć³ć³ć¹ććŖć©ćęÆč¼ććććØćć§ćć¾ććĀ
Ā
6 | NextĀ Up / ꬔåäŗå
Part 2 Building a Chat Model withĀ LangChain / LangChain ć§ćć£ććć¢ćć«ćę§ēÆ
Weāll finally implement the chat model!Ā Make sure your Python development environment is readyāVS Code or a similar IDE will work perfectly!
Part 2ć§ćÆććććććć£ććć¢ćć«ć®å®č£ ć«å „ćć¾ćļ¼VS Code ćŖć©ć使ć£ć¦ Python ć®éēŗē°å¢ćęŗåćć¦ććć¾ćććļ¼
Ā
Disclaimer / å 責äŗé
Disclaimer ā All the views and opinions in the blog are my own and is made in my personal capacity and that SAP shall not be responsible or liable for any of the contents published in this blog.
å
責äŗé
ā ę¬ććć°ć«čØč¼ćććč¦č§£ććć³ęč¦ćÆćć¹ć¦ē§åäŗŗć®ćć®ć§ćććē§ć®åäŗŗēćŖē«å “ć§ēŗäæ”ćć¦ćć¾ććSAP ćÆę¬ććć°ć®å
容ć«ć¤ćć¦äøåć®č²¬ä»»ćč² ćć¾ććć
Ā
Ā GenerativeĀ AI has leapt from research papers to daily business realityā and SAP is surfing that wave at full speed.Ā InĀ this handsāon series, Iāll show you how to spin up a custom AI agent on SAPĀ AIĀ Core in minutes, then grow it into a productionāready assetāwithout drowning in theory.ēę AI ćÆē 究室ć®å¤ćøé£ć³åŗććä»ćććøćć¹ē¾å “ć®åøøčćå”ćęæćć¦ćć¾ććSAPćÆå Øéåć§ćć®ę³¢ć«ä¹ć£ć¦ćć¾ćććć®ććć°ć·ćŖć¼ćŗć§ćÆćSAPĀ AIĀ Core ć®ę¢å®ć¢ćć«ćęéć§å¼ć³åŗććå®åć§ä½æćć AI ćØć¼ćøć§ć³ććøę”å¼µććāē§éćć³ćŗćŖć³āććå±ććć¾ććĀ šWhat Youāll Learn in This Series / ę¬ć·ćŖć¼ćŗć§å¦ć¹ćććØHow to spin up a custom AIāÆagent on SAPāÆAIĀ Core in minutesSAPĀ AIĀ Core äøć§ć«ć¹ćæć AI ćØć¼ćøć§ć³ćć āē§éā ć§åććę¹ę³Handsāon with LangChain, Google Search Tool, RAG, and StreamlitLangChainć»Google ę¤ē“¢ćć¼ć«ć»RAGć»Streamlit ć使ć£ćå®č£ Exposing the agent as a RESTĀ API and rebuilding the UI in SAPUI5/FiorićØć¼ćøć§ć³ćć RESTĀ API åććUI ć SAPUI5/Fiori ć«č¼ćęæććęé Time Commitment / å¦ēæęéEach part is designed to be completed in 10ā15Ā minutes.åē« ćÆ 10ā15 å ć§čŖććļ¼ęćåććććäŗå®ćć¦ćć¾ććĀ šŗļø Series Roadmap / é£č¼ćć¼ććććPart 0 Prologue / ćććć¼ć°Ā Part 1 Env Setup: SAP AICore & AI Launchpad / ē°å¢ę§ēÆ: SAP AI CorećØAI LaunchpadĀ [current blog / ē¾åØć®ććć°]Part 2 Building a Chat Model withĀ LangChain / LangChain ć§ćć£ććć¢ćć«ćę§ēÆPart 3 AgentĀ Tools: Integrating GoogleĀ Search / Agentćć¼ć«: GoogleĀ ę¤ē“¢ćēµ±åPart 4 RAG Basics ā : File Upload & HANA Cloud VectorEngine / RAG åŗē¤ ā : ćć”ć¤ć«ć¢ćććć¼ć & HANA Cloud ććÆćæć¼ćØć³ćøć³Part 5 RAG Basics ā”: Retriever Tool / RAG åŗē¤ ā”: Retriever Toolć®ę§ēÆPart 6 Streamlit UI Prototype / Streamlit ć§ćć£ććUIPart 7 Expose as a RESTĀ API / AI Agent ć RESTĀ API åPart 8 Rebuild theĀ UI withĀ SAPUI5 / SAPUI5 ć§ćć£ććUI ćåę§ēÆNote / 注čØSubsequent blogs will be published soon.ē¶ē·ØćÆé ę¬”å ¬éäŗå®ć§ććEnvāÆSetup: SAP AICore & AI Launchpadē°å¢ę§ēÆ: SAP AI CorećØAI Launchpad1Ā | Overview / ę¦č¦In this chapter, weāll connect to SAP AI Launchpad using the service key created for the SAP AI Core instance. Then, weāll proceed to deploy the LLM for running a chat model.ę¬ē« ć§ćÆćSAP AI Core ć¤ć³ć¹ćæć³ć¹ć«åƾćć¦ä½ęćććµć¼ćć¹ćć¼ć使ććSAP AI Launchpad ć«ę„ē¶ćć¾ćććć®å¾ććć£ććēØ LLM ćåććććć®ćććć¤ć”ć³ććå®ę½ćć¾ććĀ 2 | Prerequisites / äŗåęŗåBTP sub-account / BTP ćµćć¢ć«ć¦ć³ćSAP AI Core instance / SAP AI Core ć¤ć³ć¹ćæć³ć¹SAP AI LaunchPad Subscription / SAP AI Launchpadć®ćµćć¹ćÆćŖćć·ć§ć³Python 3.13 and pip / Python 3.13ē°å¢ & pipVSCode, BAS or any IDE / VSCodećBASćŖć©ć®IDEĀ 3Ā | StepĀ 1Ā āĀ Create a ServiceĀ Key / ServiceĀ Key ć®ä½ęCreating a serviceĀ key gives your local scripts a secure, OAuthābased passport into SAPĀ AIĀ Core.Ā Without it, Python SDK calls have no way to authenticate or discover endpoints.ćµć¼ćć¹ćć¼ćÆććć¼ć«ć«ć®ć¹ćÆćŖććć SAPĀ AIĀ Core ć«å®å Øć㤠OAuth ćć¼ć¹ć§ć¢ćÆć»ć¹ććććć® āćć¹ćć¼ćā ć§ćććććē”ććØćPython SDK ćÆćØć³ććć¤ć³ććčŖčؼę å ±ćåå¾ć§ćć¾ćććOpen your AIāÆCore instance in BTP Cockpit ā Instances &āÆSubscriptions.BTPāÆCockpit ć® Instances &āÆSubscriptions ć§ AIāÆCore ć¤ć³ć¹ćæć³ć¹ćéćć¾ććGo to the ServiceĀ Keys tab and click Create. Give it any name.ServiceĀ Keys ćæćć§ Create ććÆćŖććÆććä»»ęć®ååćå „åćć¦ä½ęćć¾ććDownload the generated JSON; youāll need it later.ēęććć JSON ććć¦ć³ćć¼ććć¦äæē®”ćć¾ććNOTE ā Resource GroupWhen calling SAPĀ AIĀ Core from Python you must set an extra environment variable named RESOURCE_GROUP in addition to the values contained in the service key. In this tutorial weāll simply point it to the builtāin default group.Python ććå¼ć³åŗćéć«ćÆćµć¼ćć¹ćć¼ć«å«ć¾ćć¦ććå¤ę°ä»„å¤ć« Resource Group ćØå¼ć°ććē°å¢å¤ę°ćåæ č¦ć§ćććę¬ćć„ć¼ććŖć¢ć«ć§ćÆę¢å®ć® defaultĀ ćęå®ćć¾ććĀ 4Ā | StepĀ 2Ā āĀ Connect AIāÆLaunchpad / AIāÆLaunchpad ćØć®ę„ē¶You can deploy an LLM (Large Language Model) directly to SAP AI Core using the service key you created earlier. However, in this guide, we’ll use SAP AI Launchpad to make the deployment process easier.To begin, you’ll need to establish a secure connection route from SAP AI Launchpad to SAP AI Core.å ć»ć©ä½ęćć SAP AI Core ć®ćµć¼ćć¹ćć¼ć使ćć°ćSAP AI Core ć«ē“ę„ć¢ćÆć»ć¹ć㦠LLMććććć¤ććććØćåÆč½ć§ćććć ććććć§ćÆććē°”åćŖę¹ę³ćØćć¦ćSAP AI Launchpadćć使ć£ć¦ćććć¤ćå®ę½ćć¾ććć¾ććÆćSAP AI Launchpad ćć SAP AI Core ćøå®å Øć«ę„ē¶ććććć®ēµč·Æćęŗåćć¾ććććFrom Subscriptions open AIāÆLaunchpad ā GoĀ toĀ Application.Subscriptions ćć AIāÆLaunchpad ā GoĀ toĀ Application ć§čµ·åćć¾ććClick Add (topāright) ā API Connection and upload the ServiceĀ Key JSON.å³äøć® Add ā API Connection ćéøć³ćServiceĀ Key JSON ćć¢ćććć¼ććć¾ććSelect the default Resource Group in the side panel and save.ćµć¤ćććć«ć§ Resource Group ć« default ćéøęćć¦äæåćć¾ććĀ 5Ā | StepĀ 3Ā āĀ Configure &Ā Deploy an LLM / ć¢ćć«čØå®ćØćććć¤On the “Configuration” screen of SAP Launchpad, you can select the type of foundation model to use. Think of this as creating a model profile that can be shared across multiple deployments.Letās go ahead and create the configuration. In this case, weāll use the default foundation_models scenario and configure it to use gapt-4o-mini.SAP Launchpad ć®ćčØå®ćē»é¢ć§ćÆć使ēØććåŗē¤ć¢ćć«ć®ēØ®é”ćŖć©ćéøć³ć¾ćććććÆćč¤ę°ć®ćććć¤ć”ć³ćć§å ±ęć§ćć āć¢ćć«ć®ćććć”ć¤ć«ā ć®ćććŖćć®ćØčćććØēč§£ććććć§ććććććć§ćÆčØå®ćä½ęćć¦ćæć¾ććććä»åćÆćććć©ć«ćć§ęä¾ććć¦ćć foundation_models ćØććć·ććŖćŖćéøęććgapt-4o-mini ćęå®ćć¾ććNavigate MLāÆOperations ā Settings and click Create.å·¦ć”ćć„ć¼ MLāÆéēØ ā čØå® ć§ ä½ę ććÆćŖććÆćć¾ććFill in: NameĀ =Ā <anyānameāyouālike>, ScenarioĀ =Ā foundation_models, VersionĀ =Ā 0.0.1, ExecutableĀ =Ā azure-openai.čØå®å <ćčŖē±ć«>, ć·ććŖćŖ foundation_models, ćć¼ćøć§ć³ 0.0.1, å®č”åÆč½ azure-openai ćå „åćć¾ććOn the parameter screen leave modelName = gpt-4o-mini, modelVersion = latest and click Next ā Create.ćć©ć”ć¼ćæē»é¢ć§ćÆ modelName = gpt-4o-mini, modelVersion = latest ććć®ć¾ć¾ć« 欔㸠ā ä½ęćYouāll land on the new Setting detail pageācheck that all fields look correct; no extra edits needed.ä½ęå¾ć«čØå®č©³ē“°ć®ē»é¢ćøé·ē§»ćć¾ććå „åå 容ć«čŖ¤ćććŖćććć£ćØē¢ŗčŖćć¾ćććļ¼åŗę¬ēć«ćć®ć¾ć¾ć§ OK ć§ćļ¼ćClickĀ CreateĀ DeploymentĀ (topāright), breeze through the wizard by clickingĀ NextĀ untilĀ Create.å³äøć®Ā ććććć¤ć”ć³ćä½ęćĀ ćę¼ććć¦ć£ć¶ć¼ććÆē¹ć«å¤ę“ććĀ ćꬔćøćĀ āĀ ćä½ęćĀ ć§é²ćć¾ććWait untilĀ Status =Ā Runningāhit the Refresh icon every few seconds. Copy theĀ DeploymentĀ ID shown in the title bar for later use.ć¹ćć¼ćæć¹ćĀ ćå®č”äøćĀ ć«ćŖćć¾ć§ćę“ę°ććæć³ć§ę°ē§ććć«ćŖćć¬ćć·ć„ćć¾ćććæć¤ćć«ćć¼ć«č”Øē¤ŗćććĀ DeploymentĀ IDĀ ćę§ćć¦ććć¾ććććBy the way, there are various LLMs available to run on SAP AI Core. You can browse them in the Generative AI Hub > Model Library section.ć”ćŖćæć«ćSAP AI Core äøć§å©ēØåÆč½ćŖ LLM ćÆćć¾ćć¾ć§ćććēę AI ćć > ć¢ćć«ć©ć¤ćć©ćŖćē»é¢ćéććØććććć®äøč¦§ć確čŖć§ćć¾ććThe Leaderboard also allows you to compare models based on accuracy, token cost, and other metrics.ć¾ćććŖć¼ćć¼ćć¼ćę©č½ć§ćÆćåć¢ćć«ć®ē²¾åŗ¦ććć¼ćÆć³ć³ć¹ććŖć©ćęÆč¼ććććØćć§ćć¾ććĀ Ā 6 | NextĀ Up / ꬔåäŗåPart 2 Building a Chat Model withĀ LangChain / LangChain ć§ćć£ććć¢ćć«ćę§ēÆWeāll finally implement the chat model!Ā Make sure your Python development environment is readyāVS Code or a similar IDE will work perfectly!Part 2ć§ćÆććććććć£ććć¢ćć«ć®å®č£ ć«å „ćć¾ćļ¼VS Code ćŖć©ć使ć£ć¦ Python ć®éēŗē°å¢ćęŗåćć¦ććć¾ćććļ¼Ā Disclaimer / å 責äŗé Disclaimer ā All the views and opinions in the blog are my own and is made in my personal capacity and that SAP shall not be responsible or liable for any of the contents published in this blog.å 責äŗé ā ę¬ććć°ć«čØč¼ćććč¦č§£ććć³ęč¦ćÆćć¹ć¦ē§åäŗŗć®ćć®ć§ćććē§ć®åäŗŗēćŖē«å “ć§ēŗäæ”ćć¦ćć¾ććSAP ćÆę¬ććć°ć®å 容ć«ć¤ćć¦äøåć®č²¬ä»»ćč² ćć¾ćććĀ Read MoreĀ Technology Blogs by SAP articlesĀ
#SAPCHANNEL