Eine Liste der unterstützten Standorte für Google-Agenten und Funktionen der Agenteninfrastruktur finden Sie unter Unterstützte Standorte für Agenten.
Google- und Partnermodelle sowie generative KI-Funktionen auf der Agent Platform werden als bestimmte regionale Endpunkte und ein globaler Endpunkt bereitgestellt. Globale Endpunkte decken die ganze Welt ab und bieten eine höhere Verfügbarkeit und Zuverlässigkeit als einzelne Regionen.
Endpunkt angeben
Für standardmäßige regionale Endpunkte (z. B. us-central) und den globalen Endpunkt:
Python
# google-cloud-aiplatform >= 1.79.0 is required
import vertexai
from vertexai.generative_models import GenerativeModel
vertexai.init(project='GOOGLE_CLOUD_PROJECT', location='GOOGLE_CLOUD_LOCATION')Python
client = genai.Client(
enterprise=True,
project='GOOGLE_CLOUD_PROJECT',
location='GOOGLE_CLOUD_LOCATION'
)REST
GOOGLE_CLOUD_PROJECT="GOOGLE_CLOUD_PROJECT"
GOOGLE_CLOUD_LOCATION="GOOGLE_CLOUD_LOCATION"
MODEL_ID="MODEL_ID"
curl -X POST \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "Content-Type: application/json" \
"https://${GOOGLE_CLOUD_LOCATION}-aiplatform.googleapis.com/v1/projects/${GOOGLE_CLOUD_PROJECT}/locations/${GOOGLE_CLOUD_LOCATION}/publishers/google/models/${MODEL_ID}:generateContent" \
-d '{
"contents": {
"role": "user",
"parts": {
"text": "Explain how AI works in a few words."
}
}
}'Multiregionale Endpunkte
Mit multiregionalen Endpunkten können Sie dafür sorgen, dass die Verarbeitung von Kundendaten durch den Dienst für maschinelles Lernen innerhalb einer bestimmten Gerichtsbarkeit erfolgt, z. B. in den USA oder der Europäischen Union.
In der folgenden Tabelle sind die Hostnamen für Multiregionen-Endpunkte aufgeführt:
| Multiregional | Standort | Hostname |
|---|---|---|
| USA | us |
https://aiplatform.us.rep.googleapis.com |
| Europäische Union | eu |
https://aiplatform.eu.rep.googleapis.com |
Python
import vertexai
from google.adk.agents import Agent
# Set the multi-region location on the client
client = vertexai.Client(
project='GOOGLE_CLOUD_PROJECT',
location='GOOGLE_CLOUD_LOCATION'
)
# The agent will use this configuration
agent = Agent(
model="gemini-3.5-flash",
name='my_agent'
)Python
from google import genai
# location must be set to 'us' or 'eu' for multi-region routing
client = genai.Client(
enterprise=True,
project='GOOGLE_CLOUD_PROJECT',
location='GOOGLE_CLOUD_LOCATION'
)REST
GOOGLE_CLOUD_PROJECT="GOOGLE_CLOUD_PROJECT"
MODEL_ID="gemini-3.5-flash"
GOOGLE_CLOUD_LOCATION="GOOGLE_CLOUD_LOCATION" # Must be 'us' or 'eu'
# Explicitly use the .rep. hostname for multi-region endpoints
API_ENDPOINT="https://aiplatform.${GOOGLE_CLOUD_LOCATION}.rep.googleapis.com"
curl -X POST \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "Content-Type: application/json" \
"${API_ENDPOINT}/v1/projects/${GOOGLE_CLOUD_PROJECT}/locations/${GOOGLE_CLOUD_LOCATION}/publishers/google/models/${MODEL_ID}:generateContent" \
-d '{
"contents": {
"role": "user",
"parts": {
"text": "Explain how AI works in a few words"
}
}
}'Private Konnektivität zu multiregionalen Endpunkten
Der privater Google-Zugriff wird für multiregionale Endpunkte nicht unterstützt. Wenn Sie versuchen, über privater Google-Zugriff eine Verbindung zu einem Multiregion-Endpunkt herzustellen, können Konnektivitätsprobleme, SSL/TLS-Handshake-Fehler oder Warnungen wegen Zertifikatsabweichungen auftreten.
Verwenden Sie Private Service Connect, um eine private Konnektivität zu multiregionalen Endpunkten herzustellen. Dies ist eine Netzwerkfunktion vonGoogle Cloud , mit der Sie privat aus Ihrem VPC-Netzwerk (Virtual Private Cloud) auf von Google verwaltete Dienste zugreifen können. Je nach den Anforderungen Ihrer Organisation in Bezug auf Compliance, Sicherheit und Traffic-Management können Sie zwischen einer direkten Endpunktverbindung und einer Load-Balancer-Architektur wählen:
- Private Service Connect-Endpunkte: Stellen eine direkte Verbindung her, indem eine interne IP-Adresse in Ihrem VPC-Netzwerk dem multiregionalen Zielendpunkt zugeordnet wird. Eine Konfigurationsanleitung finden Sie unter Über Endpunkte auf regionale und multiregionale Google APIs zugreifen.
- Private Service Connect-Backends: Verwenden Sie einen Load Balancer mit einer Private Service Connect-Netzwerk-Endpunktgruppe (NEG) für erweitertes Traffic-Management, benutzerdefinierte TLS-Zertifikate oder Richtlinienerzwingung. Eine Konfigurationsanleitung finden Sie unter Private Service Connect-Backends und Private Service Connect-Backend erstellen.
Eine vollständige Liste der unterstützten Standorte und Hostnamen finden Sie unter Regionale und multiregionale Dienstendpunkte.
Globaler Endpunkt
Wenn Sie einen globalen Endpunkt für Ihre Anfragen auswählen, kann dies die Gesamtverfügbarkeit verbessern und Fehler vom Typ „Ressourcen erschöpft“ (429) reduzieren. Verwenden Sie den globalen Endpunkt nicht, wenn Sie Anforderungen an die ML-Verarbeitung haben, da Sie nicht steuern oder wissen können, an welche Region Ihre Anfragen zur ML-Verarbeitung gesendet werden, wenn eine Anfrage gestellt wird.
Unterstützte Modelle
Welche Modelle den globalen Endpunkt unterstützen, sehen Sie auf dem Tab Global in den folgenden Tabellen:
- Standorte von Google-Modellendpunkten
- Google Cloud Endpunktstandorte für Partnermodelle
- Google Cloud Standorte für offene Modellendpunkte
Beschränkungen
Die folgenden Funktionen sind bei Verwendung des globalen Endpunkts nicht verfügbar:
- Abstimmung
- Batch-Vorhersage für Anthropic- und OpenMaaS-Modelle
- RAG-Korpus (Retrieval-Augmented Generation) (RAG-Anfragen werden unterstützt)
Die Verwendung des globalen Endpunkts mit bereitgestelltem Durchsatz ist nur für die folgenden Modelle verfügbar:
Klicken, um die unterstützten Modelle für den globalen Endpunkt bei Verwendung von Bereitgestellter Durchsatz aufzurufen
| Modell | Aktuelle unterstützte Modellversion |
|---|---|
| Gemini 4 Argon | gemini-4-argon |
| Gemini 4 Argon Cyber Permissive | gemini-4-argon-cyber-permissive |
| Gemini 3.8 Flash Cyber | gemini-3.8-flash-cyber |
| Gemini 3.8 Flash | gemini-3.8-flash |
| Gemini Omni 1.1 Flash Vorschau | gemini-omni-1.1-flash-preview |
| Gemini 3.7 Flash | gemini-3.7-flash |
| Gemini 3.6 Flash | gemini-3.6-flash |
| Gemini 3.5 Flash-Lite | gemini-3.5-flash-lite |
| Gemini 3.1 Flash-Lite Image (Nano Banana 2 Lite) | gemini-3.1-flash-lite-image |
| Gemini 3 Pro Image | gemini-3-pro-image |
| Gemini Nano Banana 2.1 | gemini-nano-banana-2.1 |
| Gemini 3.1 Flash Image | gemini-3.1-flash-image |
| Gemini 3.5 Flash | gemini-3.5-flash |
| Gemini 3.1 Flash-Lite | gemini-3.1-flash-lite |
| Gemini 3.1 Pro Vorschau | gemini-3.1-pro-preview |
| Gemini 3 Flash Vorabversion | gemini-3-flash-preview |
| Gemini 2.5 Pro | gemini-2.5-pro |
| Gemini 2.5 Flash Image | gemini-2.5-flash-image |
| Gemini 2.5 Flash | gemini-2.5-flash |
| Gemini 2.5 Flash-Lite | gemini-2.5-flash-lite |
Standorte von Google-Modellendpunkten
Google-Modelle in der Agent Platform sind für die folgenden Endpunkte verfügbar:
Global
| Global(global) | |
|---|---|
| Gemini-Modelle | |
Gemini 4 Argon
(gemini-4-argon)
|
|
Gemini 4 Argon Cyber Permissive
(gemini-4-argon-cyber-permissive)
|
|
Gemini 3.8 Live
(gemini-3.8-live)
|
|
Gemini 3.8 Flash Cyber
(gemini-3.8-flash-cyber)
|
|
Gemini 3.8 Flash
(gemini-3.8-flash)
|
|
Gemini Omni 1.1 Flash
Vorschau
(gemini-omni-1.1-flash-preview)
|
|
Gemini 3.7 Flash
(gemini-3.7-flash)
|
|
Gemini 3.6 Flash
(gemini-3.6-flash)
|
|
Gemini 3.5 Flash-Lite
(gemini-3.5-flash-lite)
|
|
Gemini Omni Flash
Vorschau
(gemini-omni-flash-preview)
|
|
Gemini 3.1 Flash-Lite Image (Nano Banana 2 Lite)
(gemini-3.1-flash-lite-image)
|
|
Gemini 3 Pro Image
(gemini-3-pro-image)
|
|
Gemini Nano Banana 2.1
(gemini-nano-banana-2.1)
|
|
Gemini 3.1 Flash Image
(gemini-3.1-flash-image)
|
|
Gemini 3.5 Flash
(gemini-3.5-flash)
|
|
Gemini 3.1 Flash-Lite
(gemini-3.1-flash-lite)
|
|
Gemini 3.1 Pro
Vorschau
(gemini-3.1-pro-preview)
|
|
Gemini 3 Flash
Vorschau
(gemini-3-flash-preview)
|
|
Gemini 2.5 Pro
(gemini-2.5-pro)
|
|
Gemini 2.5 Flash Image
(gemini-2.5-flash-image)
|
|
Gemini 2.5 Flash
(gemini-2.5-flash)
|
|
Gemini 2.5 Flash-Lite
(gemini-2.5-flash-lite)
|
|
Gemini 2.5 Flash mit Gemini Live API-nativem Audio
(gemini-live-2.5-flash-native-audio)
|
|
Virtueller Test
(virtual-try-on-001)
|
|
Gemini 3.5 Transcribe
(['gemini-3.5-transcribe', 'gemini-3.5-transcribe-live'])
|
|
Gemini 3.5 Live-Übersetzung
Vorschau
(['gemini-3.5-live-translate-preview'])
|
|
Gemini 3.8 Flash TTS
Vorschau
(gemini-3.8-flash-tts)
|
|
Gemini 3.8 Flash-Lite TTS
Vorschau
(gemini-3.8-flash-lite-tts)
|
|
| Einbettungsmodelle | |
Gemini Embedding 2
(gemini-embedding-2)
|
|
Gemini Embedding
(gemini-embedding-001)
|
|
| Einbettungen für Text | |
| Einbettungen für multimodale Konfigurationen | |
| Veo-Modelle auf der Gemini Enterprise Agent Platform | |
Veo 3 Generate
(veo-3.0-generate-001)
|
|
Veo 3 Fast Generate
(veo-3.0-fast-generate-001)
|
|
Veo 3.1 Generate
(veo-3.1-generate-001)
|
|
Veo 3.1 Fast Generate
(veo-3.1-fast-generate-001)
|
|
Veo 3.1 Lite Generate
Vorschau
(veo-3.1-lite-generate-001)
|
|
| Speech-to-Text- und Text-to-Speech-Modelle | |
Chirp 3: Transkription
(chirp_3)
|
|
| Chirp 3: HD-Stimmen | |
| Chirp 3: Instant-Custom Voice | |
Chirp 2: Transkription
(chirp_2)
|
|
Gemini 2.5 Pro TTS
(gemini-2.5-pro-tts)
|
|
Gemini 2.5 Flash TTS
(gemini-2.5-flash-tts)
|
|
Gemini 2.5 Flash Lite Preview TTS
Vorschau
(gemini-2.5-flash-lite-preview-tts)
|
|
Multiregional
| USA (mehrere Regionen)(us) | EU (mehrere Regionen)(eu) | |
|---|---|---|
| Gemini-Modelle | ||
Gemini 4 Argon
(gemini-4-argon)
|
||
Gemini 4 Argon Cyber Permissive
(gemini-4-argon-cyber-permissive)
|
||
Gemini 3.8 Live
(gemini-3.8-live)
|
||
Gemini 3.8 Flash Cyber
(gemini-3.8-flash-cyber)
|
||
Gemini 3.8 Flash
(gemini-3.8-flash)
|
||
Gemini Omni 1.1 Flash
Vorschau
(gemini-omni-1.1-flash-preview)
|
||
Gemini 3.7 Flash
(gemini-3.7-flash)
|
||
Gemini 3.6 Flash
(gemini-3.6-flash)
|
||
Gemini 3.5 Flash-Lite
(gemini-3.5-flash-lite)
|
||
Gemini Omni Flash
Vorschau
(gemini-omni-flash-preview)
|
||
Gemini 3.1 Flash-Lite Image (Nano Banana 2 Lite)
(gemini-3.1-flash-lite-image)
|
||
Gemini 3 Pro Image
(gemini-3-pro-image)
|
||
Gemini Nano Banana 2.1
(gemini-nano-banana-2.1)
|
||
Gemini 3.1 Flash Image
(gemini-3.1-flash-image)
|
||
Gemini 3.5 Flash
(gemini-3.5-flash)
|
||
Gemini 3.1 Flash-Lite
(gemini-3.1-flash-lite)
|
||
Gemini 3.1 Pro
Vorabversion
(gemini-3.1-pro-preview)
|
||
Gemini 3 Flash
Vorschau
(gemini-3-flash-preview)
|
||
Gemini 2.5 Pro
(gemini-2.5-pro)
|
||
Gemini 2.5 Flash Image
(gemini-2.5-flash-image)
|
||
Gemini 2.5 Flash
(gemini-2.5-flash)
|
||
Gemini 2.5 Flash Lite
(gemini-2.5-flash-lite)
|
||
Gemini 2.5 Flash mit nativen Audiofunktionen der Gemini Live API
(gemini-live-2.5-flash-native-audio)
|
||
Virtuelles Anprobieren
(virtual-try-on-001)
|
||
Gemini 3.5 Transcribe
(['gemini-3.5-transcribe', 'gemini-3.5-transcribe-live'])
|
||
Gemini 3.5 Live-Übersetzung
Vorschau
(['gemini-3.5-live-translate-preview'])
|
||
Gemini 3.8 Flash TTS
Vorschau
(gemini-3.8-flash-tts)
|
||
Gemini 3.8 Flash-Lite TTS
Vorschau
(gemini-3.8-flash-lite-tts)
|
||
| Einbettungsmodelle | ||
Gemini Embedding 2
(gemini-embedding-2)
|
||
Gemini Embedding
(gemini-embedding-001)
|
||
| Einbettungen für Text | ||
| Einbettungen für multimodale Konfigurationen | ||
| Veo auf Gemini Enterprise Agent Platform-Modellen | ||
Veo 3 Generate
(veo-3.0-generate-001)
|
||
Veo 3 Fast Generate
(veo-3.0-fast-generate-001)
|
||
Veo 3.1 Generate
(veo-3.1-generate-001)
|
||
Veo 3.1 Fast Generate
(veo-3.1-fast-generate-001)
|
||
Veo 3.1 Lite Generate
Vorschau
(veo-3.1-lite-generate-001)
|
||
| Speech-to-Text- und Text-to-Speech-Modelle | ||
Chirp 3: Transkription
(chirp_3)
|
||
| Chirp 3: HD-Stimmen | ||
| Chirp 3: Instant-Custom Voice | ||
Chirp 2: Transkription
(chirp_2)
|
||
Gemini 2.5 Pro TTS
(gemini-2.5-pro-tts)
|
||
Gemini 2.5 Flash TTS
(gemini-2.5-flash-tts)
|
||
Gemini 2.5 Flash Lite Preview TTS
Vorschau
(gemini-2.5-flash-lite-preview-tts)
|
||
USA
| Oregon(us-west1) | Las Vegas(us-west4) | Iowa(us-central1) | South Carolina(us-east1) | N. Virginia(us-east4) | Columbus(us-east5) | Dallas(us-south1) | |
|---|---|---|---|---|---|---|---|
| Gemini-Modelle | |||||||
Gemini 4 Argon
(gemini-4-argon)
|
|||||||
Gemini 4 Argon Cyber Permissive
(gemini-4-argon-cyber-permissive)
|
|||||||
Gemini 3.8 Live
(gemini-3.8-live)
|
|||||||
Gemini 3.8 Flash Cyber
(gemini-3.8-flash-cyber)
|
|||||||
Gemini 3.8 Flash
(gemini-3.8-flash)
|
|||||||
Gemini Omni 1.1 Flash
Vorschau
(gemini-omni-1.1-flash-preview)
|
|||||||
Gemini 3.7 Flash
(gemini-3.7-flash)
|
|||||||
Gemini 3.6 Flash
(gemini-3.6-flash)
|
|||||||
Gemini 3.5 Flash-Lite
(gemini-3.5-flash-lite)
|
|||||||
Gemini Omni Flash
Vorschau
(gemini-omni-flash-preview)
|
|||||||
Gemini 3.1 Flash-Lite Image (Nano Banana 2 Lite)
(gemini-3.1-flash-lite-image)
|
|||||||
Gemini 3 Pro Image
(gemini-3-pro-image)
|
|||||||
Gemini Nano Banana 2.1
(gemini-nano-banana-2.1)
|
|||||||
Gemini 3.1 Flash Image
(gemini-3.1-flash-image)
|
|||||||
Gemini 3.5 Flash
(gemini-3.5-flash)
|
|||||||
Gemini 3.1 Flash-Lite
(gemini-3.1-flash-lite)
|
|||||||
Gemini 3.1 Pro
Vorschau
(gemini-3.1-pro-preview)
|
|||||||
Gemini 3 Flash
Vorschau
(gemini-3-flash-preview)
|
|||||||
Gemini 2.5 Pro
(gemini-2.5-pro)
|
|||||||
Gemini 2.5 Flash Image
(gemini-2.5-flash-image)
|
|||||||
Gemini 2.5 Flash
(gemini-2.5-flash)
|
|||||||
Gemini 2.5 Flash-Lite
(gemini-2.5-flash-lite)
|
|||||||
Gemini 2.5 Flash mit nativen Audiofunktionen der Gemini Live API
(gemini-live-2.5-flash-native-audio)
|
|||||||
Virtueller Test
(virtual-try-on-001)
|
|||||||
Gemini 3.5 Transcribe
(['gemini-3.5-transcribe', 'gemini-3.5-transcribe-live'])
|
|||||||
Gemini 3.5 Live Translate
Vorschau
(['gemini-3.5-live-translate-preview'])
|
|||||||
Gemini 3.8 Flash TTS
Vorschau
(gemini-3.8-flash-tts)
|
|||||||
Gemini 3.8 Flash-Lite TTS
Vorschau
(gemini-3.8-flash-lite-tts)
|
|||||||
| Einbettungsmodelle | |||||||
Gemini Embedding 2
(gemini-embedding-2)
|
|||||||
Gemini Embedding
(gemini-embedding-001)
|
|||||||
| Einbettungen für Text | |||||||
| Einbettungen für multimodale Konfigurationen | |||||||
| Veo-Modelle auf der Gemini Enterprise Agent Platform | |||||||
Veo 3 Generate
(veo-3.0-generate-001)
|
|||||||
Veo 3 Fast Generate
(veo-3.0-fast-generate-001)
|
|||||||
Veo 3.1 Generate
(veo-3.1-generate-001)
|
|||||||
Veo 3.1 Fast Generate
(veo-3.1-fast-generate-001)
|
|||||||
Veo 3.1 Lite Generate
Vorschau
(veo-3.1-lite-generate-001)
|
|||||||
| Speech-to-Text- und Text-to-Speech-Modelle | |||||||
Chirp 3: Transkription
(chirp_3)
|
|||||||
| Chirp 3: HD-Stimmen | |||||||
| Chirp 3: Instant-Custom Voice | |||||||
Chirp 2: Transkription
(chirp_2)
|
|||||||
Gemini 2.5 Pro TTS
(gemini-2.5-pro-tts)
|
|||||||
Gemini 2.5 Flash TTS
(gemini-2.5-flash-tts)
|
|||||||
Gemini 2.5 Flash Lite Preview TTS
Vorschau
(gemini-2.5-flash-lite-preview-tts)
|
|||||||
Nord- und Südamerika
| Montreal(northamerica-northeast1) | São Paulo(southamerica-east1) | |
|---|---|---|
| Gemini-Modelle | ||
Gemini 4 Argon
(gemini-4-argon)
|
||
Gemini 4 Argon Cyber Permissive
(gemini-4-argon-cyber-permissive)
|
||
Gemini 3.8 Live
(gemini-3.8-live)
|
||
Gemini 3.8 Flash Cyber
(gemini-3.8-flash-cyber)
|
||
Gemini 3.8 Flash
(gemini-3.8-flash)
|
||
Gemini Omni 1.1 Flash
Vorschau
(gemini-omni-1.1-flash-preview)
|
||
Gemini 3.7 Flash
(gemini-3.7-flash)
|
||
Gemini 3.6 Flash
(gemini-3.6-flash)
|
||
Gemini 3.5 Flash-Lite
(gemini-3.5-flash-lite)
|
||
Gemini Omni Flash
Vorschau
(gemini-omni-flash-preview)
|
||
Gemini 3.1 Flash-Lite Image (Nano Banana 2 Lite)
(gemini-3.1-flash-lite-image)
|
||
Gemini 3 Pro Image
(gemini-3-pro-image)
|
||
Gemini Nano Banana 2.1
(gemini-nano-banana-2.1)
|
||
Gemini 3.1 Flash Image
(gemini-3.1-flash-image)
|
||
Gemini 3.5 Flash
(gemini-3.5-flash)
|
||
Gemini 3.1 Flash-Lite
(gemini-3.1-flash-lite)
|
||
Gemini 3.1 Pro
Vorschau
(gemini-3.1-pro-preview)
|
||
Gemini 3 Flash
Vorschau
(gemini-3-flash-preview)
|
||
Gemini 2.5 Pro
(gemini-2.5-pro)
|
||
Gemini 2.5 Flash Image
(gemini-2.5-flash-image)
|
||
Gemini 2.5 Flash
(gemini-2.5-flash)
|
||
Gemini 2.5 Flash-Lite
(gemini-2.5-flash-lite)
|
||
Gemini 2.5 Flash mit Gemini Live API-nativem Audio
(gemini-live-2.5-flash-native-audio)
|
||
Virtuelles Anprobieren
(virtual-try-on-001)
|
||
Gemini 3.5 Transcribe
(['gemini-3.5-transcribe', 'gemini-3.5-transcribe-live'])
|
||
Gemini 3.5 Live Translate
Vorschau
(['gemini-3.5-live-translate-preview'])
|
||
Gemini 3.8 Flash TTS
Vorschau
(gemini-3.8-flash-tts)
|
||
Gemini 3.8 Flash-Lite TTS
Vorschau
(gemini-3.8-flash-lite-tts)
|
||
| Einbettungsmodelle | ||
Gemini Embedding 2
(gemini-embedding-2)
|
||
Gemini Embedding
(gemini-embedding-001)
|
||
| Einbettungen für Text | ||
| Einbettungen für multimodale Konfigurationen | ||
| Veo-Modelle auf der Gemini Enterprise Agent Platform | ||
Veo 3 Generate
(veo-3.0-generate-001)
|
||
Veo 3 Fast Generate
(veo-3.0-fast-generate-001)
|
||
Veo 3.1 Generate
(veo-3.1-generate-001)
|
||
Veo 3.1 Fast Generate
(veo-3.1-fast-generate-001)
|
||
Veo 3.1 Lite Generate
Vorschau
(veo-3.1-lite-generate-001)
|
||
| Speech-to-Text- und Text-to-Speech-Modelle | ||
Chirp 3: Transkription
(chirp_3)
|
||
| Chirp 3: HD-Stimmen | ||
| Chirp 3: Instant-Custom Voice | ||
Chirp 2: Transkription
(chirp_2)
|
||
Gemini 2.5 Pro TTS
(gemini-2.5-pro-tts)
|
||
Gemini 2.5 Flash TTS
(gemini-2.5-flash-tts)
|
||
Gemini 2.5 Flash Lite Preview TTS
Vorschau
(gemini-2.5-flash-lite-preview-tts)
|
||
Europa
| London(europe-west2) | Belgien(europe-west1) | Niederlande(europe-west4) | Zürich(europe-west6) | Frankfurt(europe-west3) | Finnland(europe-north1) | Warschau(europe-central2) | Mailand(europe-west8) | Madrid(europe-southwest1) | Paris(europe-west9) | |
|---|---|---|---|---|---|---|---|---|---|---|
| Gemini-Modelle | ||||||||||
Gemini 4 Argon
(gemini-4-argon)
|
||||||||||
Gemini 4 Argon Cyber Permissive
(gemini-4-argon-cyber-permissive)
|
||||||||||
Gemini 3.8 Live
(gemini-3.8-live)
|
||||||||||
Gemini 3.8 Flash Cyber
(gemini-3.8-flash-cyber)
|
||||||||||
Gemini 3.8 Flash
(gemini-3.8-flash)
|
||||||||||
Gemini Omni 1.1 Flash
Vorschau
(gemini-omni-1.1-flash-preview)
|
||||||||||
Gemini 3.7 Flash
(gemini-3.7-flash)
|
||||||||||
Gemini 3.6 Flash
(gemini-3.6-flash)
|
||||||||||
Gemini 3.5 Flash-Lite
(gemini-3.5-flash-lite)
|
||||||||||
Gemini Omni Flash
Vorschau
(gemini-omni-flash-preview)
|
||||||||||
Gemini 3.1 Flash-Lite Image (Nano Banana 2 Lite)
(gemini-3.1-flash-lite-image)
|
||||||||||
Gemini 3 Pro Image
(gemini-3-pro-image)
|
||||||||||
Gemini Nano Banana 2.1
(gemini-nano-banana-2.1)
|
||||||||||
Gemini 3.1 Flash Image
(gemini-3.1-flash-image)
|
||||||||||
Gemini 3.5 Flash
(gemini-3.5-flash)
|
||||||||||
Gemini 3.1 Flash-Lite
(gemini-3.1-flash-lite)
|
||||||||||
Gemini 3.1 Pro
Vorschau
(gemini-3.1-pro-preview)
|
||||||||||
Gemini 3 Flash
Vorschau
(gemini-3-flash-preview)
|
||||||||||
Gemini 2.5 Pro
(gemini-2.5-pro)
|
||||||||||
Gemini 2.5 Flash Image
(gemini-2.5-flash-image)
|
||||||||||
Gemini 2.5 Flash
(gemini-2.5-flash)
|
||||||||||
Gemini 2.5 Flash-Lite
(gemini-2.5-flash-lite)
|
||||||||||
Gemini 2.5 Flash mit Gemini Live API-nativem Audio
(gemini-live-2.5-flash-native-audio)
|
||||||||||
Virtuelles Anprobieren
(virtual-try-on-001)
|
||||||||||
Gemini 3.5 Transcribe
(['gemini-3.5-transcribe', 'gemini-3.5-transcribe-live'])
|
||||||||||
Gemini 3.5 Live Translate
Vorschau
(['gemini-3.5-live-translate-preview'])
|
||||||||||
Gemini 3.8 Flash TTS
Vorschau
(gemini-3.8-flash-tts)
|
||||||||||
Gemini 3.8 Flash-Lite TTS
Vorschau
(gemini-3.8-flash-lite-tts)
|
||||||||||
| Einbettungsmodelle | ||||||||||
Gemini Embedding 2
(gemini-embedding-2)
|
||||||||||
Gemini Embedding
(gemini-embedding-001)
|
||||||||||
| Einbettungen für Text | ||||||||||
| Einbettungen für multimodale Konfigurationen | ||||||||||
| Veo auf Gemini Enterprise Agent Platform-Modellen | ||||||||||
Veo 3 Generate
(veo-3.0-generate-001)
|
||||||||||
Veo 3 Fast Generate
(veo-3.0-fast-generate-001)
|
||||||||||
Veo 3.1 Generate
(veo-3.1-generate-001)
|
||||||||||
Veo 3.1 Fast Generate
(veo-3.1-fast-generate-001)
|
||||||||||
Veo 3.1 Lite Generate
Vorschau
(veo-3.1-lite-generate-001)
|
||||||||||
| Speech-to-Text- und Text-to-Speech-Modelle | ||||||||||
Chirp 3: Transkription
(chirp_3)
|
||||||||||
| Chirp 3: HD-Stimmen | ||||||||||
| Chirp 3: Instant-Custom Voice | ||||||||||
Chirp 2: Transkription
(chirp_2)
|
||||||||||
Gemini 2.5 Pro TTS
(gemini-2.5-pro-tts)
|
||||||||||
Gemini 2.5 Flash TTS
(gemini-2.5-flash-tts)
|
||||||||||
Gemini 2.5 Flash Lite Preview TTS
Vorschau
(gemini-2.5-flash-lite-preview-tts)
|
||||||||||
Asiatisch-pazifischer Raum
| Mumbai(asia-south1) | Singapur(asia-southeast1) | Hongkong(asia-east2) | Taiwan(asia-east1) | Tokio(asia-northeast1) | Sydney(australia-southeast1) | Seoul(asia-northeast3) | |
|---|---|---|---|---|---|---|---|
| Gemini-Modelle | |||||||
Gemini 4 Argon
(gemini-4-argon)
|
|||||||
Gemini 4 Argon Cyber Permissive
(gemini-4-argon-cyber-permissive)
|
|||||||
Gemini 3.8 Live
(gemini-3.8-live)
|
|||||||
Gemini 3.8 Flash Cyber
(gemini-3.8-flash-cyber)
|
|||||||
Gemini 3.8 Flash
(gemini-3.8-flash)
|
|||||||
Gemini Omni 1.1 Flash
Vorschau
(gemini-omni-1.1-flash-preview)
|
|||||||
Gemini 3.7 Flash
(gemini-3.7-flash)
|
|||||||
Gemini 3.6 Flash
(gemini-3.6-flash)
|
|||||||
Gemini 3.5 Flash-Lite
(gemini-3.5-flash-lite)
|
|||||||
Gemini Omni Flash
Vorschau
(gemini-omni-flash-preview)
|
|||||||
Gemini 3.1 Flash-Lite Image (Nano Banana 2 Lite)
(gemini-3.1-flash-lite-image)
|
|||||||
Gemini 3 Pro Image
(gemini-3-pro-image)
|
|||||||
Gemini Nano Banana 2.1
(gemini-nano-banana-2.1)
|
|||||||
Gemini 3.1 Flash Image
(gemini-3.1-flash-image)
|
|||||||
Gemini 3.5 Flash
(gemini-3.5-flash)
|
|||||||
Gemini 3.1 Flash-Lite
(gemini-3.1-flash-lite)
|
|||||||
Gemini 3.1 Pro
Vorabversion
(gemini-3.1-pro-preview)
|
|||||||
Gemini 3 Flash
Vorschau
(gemini-3-flash-preview)
|
|||||||
Gemini 2.5 Pro
(gemini-2.5-pro)
|
|||||||
Gemini 2.5 Flash Image
(gemini-2.5-flash-image)
|
|||||||
Gemini 2.5 Flash
(gemini-2.5-flash)
|
|||||||
Gemini 2.5 Flash-Lite
(gemini-2.5-flash-lite)
|
|||||||
Gemini 2.5 Flash mit nativen Audiofunktionen der Gemini Live API
(gemini-live-2.5-flash-native-audio)
|
|||||||
Virtuelles Anprobieren
(virtual-try-on-001)
|
|||||||
Gemini 3.5 Transcribe
(['gemini-3.5-transcribe', 'gemini-3.5-transcribe-live'])
|
|||||||
Gemini 3.5 Live-Übersetzung
Vorschau
(['gemini-3.5-live-translate-preview'])
|
|||||||
Gemini 3.8 Flash TTS
Vorschau
(gemini-3.8-flash-tts)
|
|||||||
Gemini 3.8 Flash-Lite TTS
Vorschau
(gemini-3.8-flash-lite-tts)
|
|||||||
| Einbettungsmodelle | |||||||
Gemini Embedding 2
(gemini-embedding-2)
|
|||||||
Gemini Embedding
(gemini-embedding-001)
|
|||||||
| Einbettungen für Text | |||||||
| Einbettungen für multimodale Konfigurationen | |||||||
| Veo-Modelle auf der Gemini Enterprise Agent Platform | |||||||
Veo 3 Generate
(veo-3.0-generate-001)
|
|||||||
Veo 3 Fast Generate
(veo-3.0-fast-generate-001)
|
|||||||
Veo 3.1 Generate
(veo-3.1-generate-001)
|
|||||||
Veo 3.1 Fast Generate
(veo-3.1-fast-generate-001)
|
|||||||
Veo 3.1 Lite Generate
Vorschau
(veo-3.1-lite-generate-001)
|
|||||||
| Speech-to-Text- und Text-to-Speech-Modelle | |||||||
Chirp 3: Transkription
(chirp_3)
|
|||||||
| Chirp 3: HD-Stimmen | |||||||
| Chirp 3: Instant-Custom Voice | |||||||
Chirp 2: Transkription
(chirp_2)
|
|||||||
Gemini 2.5 Pro TTS
(gemini-2.5-pro-tts)
|
|||||||
Gemini 2.5 Flash TTS
(gemini-2.5-flash-tts)
|
|||||||
Gemini 2.5 Flash Lite Preview TTS
Vorschau
(gemini-2.5-flash-lite-preview-tts)
|
|||||||
Naher Osten
| Tel Aviv(me-west1) | Doha(me-central1) | Dammam(me-central2) | |
|---|---|---|---|
| Gemini-Modelle | |||
Gemini 4 Argon
(gemini-4-argon)
|
|||
Gemini 4 Argon Cyber Permissive
(gemini-4-argon-cyber-permissive)
|
|||
Gemini 3.8 Live
(gemini-3.8-live)
|
|||
Gemini 3.8 Flash Cyber
(gemini-3.8-flash-cyber)
|
|||
Gemini 3.8 Flash
(gemini-3.8-flash)
|
|||
Gemini Omni 1.1 Flash
Vorschau
(gemini-omni-1.1-flash-preview)
|
|||
Gemini 3.7 Flash
(gemini-3.7-flash)
|
|||
Gemini 3.6 Flash
(gemini-3.6-flash)
|
|||
Gemini 3.5 Flash-Lite
(gemini-3.5-flash-lite)
|
|||
Gemini Omni Flash
Vorschau
(gemini-omni-flash-preview)
|
|||
Gemini 3.1 Flash-Lite Image (Nano Banana 2 Lite)
(gemini-3.1-flash-lite-image)
|
|||
Gemini 3 Pro Image
(gemini-3-pro-image)
|
|||
Gemini Nano Banana 2.1
(gemini-nano-banana-2.1)
|
|||
Gemini 3.1 Flash Image
(gemini-3.1-flash-image)
|
|||
Gemini 3.5 Flash
(gemini-3.5-flash)
|
|||
Gemini 3.1 Flash-Lite
(gemini-3.1-flash-lite)
|
|||
Gemini 3.1 Pro
Vorschau
(gemini-3.1-pro-preview)
|
|||
Gemini 3 Flash
Vorschau
(gemini-3-flash-preview)
|
|||
Gemini 2.5 Pro
(gemini-2.5-pro)
|
|||
Gemini 2.5 Flash Image
(gemini-2.5-flash-image)
|
|||
Gemini 2.5 Flash
(gemini-2.5-flash)
|
|||
Gemini 2.5 Flash-Lite
(gemini-2.5-flash-lite)
|
|||
Gemini 2.5 Flash mit nativer Audioausgabe der Gemini Live API
(gemini-live-2.5-flash-native-audio)
|
|||
Virtuelles Anprobieren
(virtual-try-on-001)
|
|||
Gemini 3.5 Transcribe
(['gemini-3.5-transcribe', 'gemini-3.5-transcribe-live'])
|
|||
Gemini 3.5 Live-Übersetzung
Vorschau
(['gemini-3.5-live-translate-preview'])
|
|||
Gemini 3.8 Flash TTS
Vorschau
(gemini-3.8-flash-tts)
|
|||
Gemini 3.8 Flash-Lite TTS
Vorschau
(gemini-3.8-flash-lite-tts)
|
|||
| Einbettungsmodelle | |||
Gemini Embedding 2
(gemini-embedding-2)
|
|||
Gemini Embedding
(gemini-embedding-001)
|
|||
| Einbettungen für Text | |||
| Einbettungen für multimodale Konfigurationen | |||
| Veo auf Gemini Enterprise Agent Platform-Modellen | |||
Veo 3 Generate
(veo-3.0-generate-001)
|
|||
Veo 3 Fast Generate
(veo-3.0-fast-generate-001)
|
|||
Veo 3.1 Generate
(veo-3.1-generate-001)
|
|||
Veo 3.1 Fast Generate
(veo-3.1-fast-generate-001)
|
|||
Veo 3.1 Lite Generate
Vorschau
(veo-3.1-lite-generate-001)
|
|||
| Speech-to-Text- und Text-to-Speech-Modelle | |||
Chirp 3: Transkription
(chirp_3)
|
|||
| Chirp 3: HD-Stimmen | |||
| Chirp 3: Instant-Custom Voice | |||
Chirp 2: Transkription
(chirp_2)
|
|||
Gemini 2.5 Pro TTS
(gemini-2.5-pro-tts)
|
|||
Gemini 2.5 Flash TTS
(gemini-2.5-flash-tts)
|
|||
Gemini 2.5 Flash Lite Preview TTS
Vorschau
(gemini-2.5-flash-lite-preview-tts)
|
|||
Google Cloud Standorte für Partnermodell-Endpunkte
Google verarbeitet Anfragen aus der von Ihnen angegebenen Region. Für einige Modelle bietet Google auch einen globalen Endpunkt an, um die allgemeine Verfügbarkeit zu verbessern und die Fehlerraten zu senken. Der globale Endpunkt kann andere Kontingente als der regionale Endpunkt haben und unterstützt keine Anforderungen an den Datenstandort. Weitere Informationen finden Sie unter Plattformpartnermodelle für die Agent Platform für MaaS.
Partnermodell-Endpunkte für die Agent Platform sind in den folgenden Regionen verfügbar:
Global
| Global(global) | |
|---|---|
| Anthropic-Modelle | |
| Claude Haiku 5.5 in Google Cloud | |
| Claude Sonnet 5.5 in Google Cloud | |
| Claude Sonnet 5 in Google Cloud | |
| Claude Opus 5.5 in Google Cloud | |
| Claude Opus 5 in Google Cloud | |
| Claude Fable 5.1 in Google Cloud | |
| Claude Fable 5 in Google Cloud | |
| Claude Opus 4.8 in Google Cloud | |
| Claude Opus 4.7 in Google Cloud | |
| Claude Opus 4.6 in Google Cloud | |
| Claude Opus 4.5 in Google Cloud | |
| Claude Sonnet 4.6 in Google Cloud | |
| Claude Sonnet 4.5 in Google Cloud | |
| Claude Opus 4.1 in Google Cloud | |
| Claude Haiku 4.5 in Google Cloud | |
| Claude Opus 4 in Google Cloud | |
| Claude Sonnet 4 in Google Cloud | |
| Claude 3.7 Sonnet von Anthropic in Google Cloud | |
| Anthropic Claude 3.5 Haiku in Google Cloud | |
| Claude 3 Haiku von Anthropic in Google Cloud (eingestellt) | |
| Mistral-Modelle | |
| Mistral Medium 3 | |
| Mistral OCR (25.05) | |
| Mistral Small 3.1 (25.03) | |
| Mistral Large (24.07) | |
| Codestral 2 | |
| Codestral (24.05) | |
| Grok-Modelle | |
| Grok 4.6 | |
| Grok 4.20 (Reasoning) | |
| Grok 4.20 (ohne Begründung) | |
| Grok 4.1 Fast (Reasoning) | |
| Grok 4.1 Fast (ohne Reasoning) | |
Multiregional
| USA (mehrere Regionen)(us) | EU (mehrere Regionen)(eu) | |
|---|---|---|
| Anthropic-Modelle | ||
| Claude Haiku 5.5 in Google Cloud | ||
| Claude Sonnet 5.5 in Google Cloud | ||
| Claude Sonnet 5 in Google Cloud | ||
| Claude Opus 5.5 in Google Cloud | ||
| Claude Opus 5 in Google Cloud | ||
| Claude Fable 5.1 in Google Cloud | ||
| Claude Fable 5 in Google Cloud | ||
| Claude Opus 4.8 in Google Cloud | ||
| Claude Opus 4.7 in Google Cloud | ||
| Claude Opus 4.6 in Google Cloud | ||
| Claude Opus 4.5 in Google Cloud | ||
| Claude Sonnet 4.6 in Google Cloud | ||
| Claude Sonnet 4.5 in Google Cloud | ||
| Claude Opus 4.1 in Google Cloud | ||
| Claude Haiku 4.5 in Google Cloud | ||
| Claude Opus 4 in Google Cloud | ||
| Claude Sonnet 4 in Google Cloud | ||
| Claude 3.7 Sonnet von Anthropic in Google Cloud | ||
| Anthropic Claude 3.5 Haiku in Google Cloud | ||
| Claude 3 Haiku von Anthropic in Google Cloud (eingestellt) | ||
| Mistral-Modelle | ||
| Mistral Medium 3 | ||
| Mistral OCR (25.05) | ||
| Mistral Small 3.1 (25.03) | ||
| Mistral Large (24.07) | ||
| Codestral 2 | ||
| Codestral (24.05) | ||
| Grok-Modelle | ||
| Grok 4.6 | ||
| Grok 4.20 (Reasoning) | ||
| Grok 4.20 (ohne Begründung) | ||
| Grok 4.1 Fast (Reasoning) | ||
| Grok 4.1 Fast (ohne Begründung) | ||
USA
| Oregon(us-west1) | Las Vegas(us-west4) | Iowa(us-central1) | South Carolina(us-east1) | N. Virginia(us-east4) | Columbus(us-east5) | Dallas(us-south1) | |
|---|---|---|---|---|---|---|---|
| Anthropic-Modelle | |||||||
| Claude Haiku 5.5 in Google Cloud | |||||||
| Claude Sonnet 5.5 in Google Cloud | |||||||
| Claude Sonnet 5 in Google Cloud | |||||||
| Claude Opus 5.5 in Google Cloud | |||||||
| Claude Opus 5 in Google Cloud | |||||||
| Claude Fable 5.1 in Google Cloud | |||||||
| Claude Fable 5 in Google Cloud | |||||||
| Claude Opus 4.8 in Google Cloud | |||||||
| Claude Opus 4.7 in Google Cloud | |||||||
| Claude Opus 4.6 in Google Cloud | |||||||
| Claude Opus 4.5 in Google Cloud | |||||||
| Claude Sonnet 4.6 in Google Cloud | |||||||
| Claude Sonnet 4.5 in Google Cloud | |||||||
| Claude Opus 4.1 in Google Cloud | |||||||
| Claude Haiku 4.5 in Google Cloud | |||||||
| Claude Opus 4 in Google Cloud | |||||||
| Claude Sonnet 4 in Google Cloud | |||||||
| Claude 3.7 Sonnet von Anthropic in Google Cloud | |||||||
| Claude 3.5 Haiku von Anthropic in Google Cloud | |||||||
| Claude 3 Haiku von Anthropic in Google Cloud (eingestellt) | |||||||
| Mistral-Modelle | |||||||
| Mistral Medium 3 | |||||||
| Mistral OCR (25.05) | |||||||
| Mistral Small 3.1 (25.03) | |||||||
| Mistral Large (24.07) | |||||||
| Codestral 2 | |||||||
| Codestral (24.05) | |||||||
| Grok-Modelle | |||||||
| Grok 4.6 | |||||||
| Grok 4.20 (Reasoning) | |||||||
| Grok 4.20 (ohne Begründung) | |||||||
| Grok 4.1 Fast (Reasoning) | |||||||
| Grok 4.1 Fast (ohne Reasoning) | |||||||
Nord- und Südamerika
| Montreal(northamerica-northeast1) | São Paulo(southamerica-east1) | |
|---|---|---|
| Anthropic-Modelle | ||
| Claude Haiku 5.5 in Google Cloud | ||
| Claude Sonnet 5.5 in Google Cloud | ||
| Claude Sonnet 5 in Google Cloud | ||
| Claude Opus 5.5 in Google Cloud | ||
| Claude Opus 5 in Google Cloud | ||
| Claude Fable 5.1 in Google Cloud | ||
| Claude Fable 5 in Google Cloud | ||
| Claude Opus 4.8 in Google Cloud | ||
| Claude Opus 4.7 in Google Cloud | ||
| Claude Opus 4.6 in Google Cloud | ||
| Claude Opus 4.5 in Google Cloud | ||
| Claude Sonnet 4.6 in Google Cloud | ||
| Claude Sonnet 4.5 in Google Cloud | ||
| Claude Opus 4.1 in Google Cloud | ||
| Claude Haiku 4.5 in Google Cloud | ||
| Claude Opus 4 in Google Cloud | ||
| Claude Sonnet 4 in Google Cloud | ||
| Claude 3.7 Sonnet von Anthropic in Google Cloud | ||
| Anthropic Claude 3.5 Haiku in Google Cloud | ||
| Claude 3 Haiku von Anthropic in Google Cloud (eingestellt) | ||
| Mistral-Modelle | ||
| Mistral Medium 3 | ||
| Mistral OCR (25.05) | ||
| Mistral Small 3.1 (25.03) | ||
| Mistral Large (24.07) | ||
| Codestral 2 | ||
| Codestral (24.05) | ||
| Grok-Modelle | ||
| Grok 4.6 | ||
| Grok 4.20 (Reasoning) | ||
| Grok 4.20 (ohne Begründung) | ||
| Grok 4.1 Fast (Reasoning) | ||
| Grok 4.1 Fast (ohne Begründung) | ||
Europa
| London(europe-west2) | Belgien(europe-west1) | Niederlande(europe-west4) | Zürich(europe-west6) | Frankfurt(europe-west3) | Finnland(europe-north1) | Warschau(europe-central2) | Mailand(europe-west8) | Madrid(europe-southwest1) | Paris(europe-west9) | |
|---|---|---|---|---|---|---|---|---|---|---|
| Anthropic-Modelle | ||||||||||
| Claude Haiku 5.5 in Google Cloud | ||||||||||
| Claude Sonnet 5.5 in Google Cloud | ||||||||||
| Claude Sonnet 5 in Google Cloud | ||||||||||
| Claude Opus 5.5 in Google Cloud | ||||||||||
| Claude Opus 5 in Google Cloud | ||||||||||
| Claude Fable 5.1 in Google Cloud | ||||||||||
| Claude Fable 5 in Google Cloud | ||||||||||
| Claude Opus 4.8 in Google Cloud | ||||||||||
| Claude Opus 4.7 in Google Cloud | ||||||||||
| Claude Opus 4.6 in Google Cloud | ||||||||||
| Claude Opus 4.5 in Google Cloud | ||||||||||
| Claude Sonnet 4.6 in Google Cloud | ||||||||||
| Claude Sonnet 4.5 in Google Cloud | ||||||||||
| Claude Opus 4.1 in Google Cloud | ||||||||||
| Claude Haiku 4.5 in Google Cloud | ||||||||||
| Claude Opus 4 in Google Cloud | ||||||||||
| Claude Sonnet 4 in Google Cloud | ||||||||||
| Claude 3.7 Sonnet von Anthropic in Google Cloud | ||||||||||
| Claude 3.5 Haiku von Anthropic in Google Cloud | ||||||||||
| Claude 3 Haiku von Anthropic in Google Cloud (eingestellt) | ||||||||||
| Mistral-Modelle | ||||||||||
| Mistral Medium 3 | ||||||||||
| Mistral OCR (25.05) | ||||||||||
| Mistral Small 3.1 (25.03) | ||||||||||
| Mistral Large (24.07) | ||||||||||
| Codestral 2 | ||||||||||
| Codestral (24.05) | ||||||||||
| Grok-Modelle | ||||||||||
| Grok 4.6 | ||||||||||
| Grok 4.20 (Reasoning) | ||||||||||
| Grok 4.20 (ohne Begründung) | ||||||||||
| Grok 4.1 Fast (Reasoning) | ||||||||||
| Grok 4.1 Fast (ohne Reasoning) | ||||||||||
Asiatisch-pazifischer Raum
| Mumbai(asia-south1) | Singapur(asia-southeast1) | Hongkong(asia-east2) | Taiwan(asia-east1) | Tokio(asia-northeast1) | Sydney(australia-southeast1) | Seoul(asia-northeast3) | |
|---|---|---|---|---|---|---|---|
| Anthropic-Modelle | |||||||
| Claude Haiku 5.5 in Google Cloud | |||||||
| Claude Sonnet 5.5 in Google Cloud | |||||||
| Claude Sonnet 5 in Google Cloud | |||||||
| Claude Opus 5.5 in Google Cloud | |||||||
| Claude Opus 5 in Google Cloud | |||||||
| Claude Fable 5.1 in Google Cloud | |||||||
| Claude Fable 5 in Google Cloud | |||||||
| Claude Opus 4.8 in Google Cloud | |||||||
| Claude Opus 4.7 in Google Cloud | |||||||
| Claude Opus 4.6 in Google Cloud | |||||||
| Claude Opus 4.5 in Google Cloud | |||||||
| Claude Sonnet 4.6 in Google Cloud | |||||||
| Claude Sonnet 4.5 in Google Cloud | |||||||
| Claude Opus 4.1 in Google Cloud | |||||||
| Claude Haiku 4.5 in Google Cloud | |||||||
| Claude Opus 4 in Google Cloud | |||||||
| Claude Sonnet 4 in Google Cloud | |||||||
| Claude 3.7 Sonnet von Anthropic in Google Cloud | |||||||
| Anthropic Claude 3.5 Haiku in Google Cloud | |||||||
| Claude 3 Haiku von Anthropic in Google Cloud (eingestellt) | |||||||
| Mistral-Modelle | |||||||
| Mistral Medium 3 | |||||||
| Mistral OCR (25.05) | |||||||
| Mistral Small 3.1 (25.03) | |||||||
| Mistral Large (24.07) | |||||||
| Codestral 2 | |||||||
| Codestral (24.05) | |||||||
| Grok-Modelle | |||||||
| Grok 4.6 | |||||||
| Grok 4.20 (Reasoning) | |||||||
| Grok 4.20 (ohne Begründung) | |||||||
| Grok 4.1 Fast (Reasoning) | |||||||
| Grok 4.1 Fast (ohne Reasoning) | |||||||
Naher Osten
| Tel Aviv(me-west1) | Doha(me-central1) | Dammam(me-central2) | |
|---|---|---|---|
| Anthropic-Modelle | |||
| Claude Haiku 5.5 in Google Cloud | |||
| Claude Sonnet 5.5 in Google Cloud | |||
| Claude Sonnet 5 in Google Cloud | |||
| Claude Opus 5.5 in Google Cloud | |||
| Claude Opus 5 in Google Cloud | |||
| Claude Fable 5.1 in Google Cloud | |||
| Claude Fable 5 in Google Cloud | |||
| Claude Opus 4.8 in Google Cloud | |||
| Claude Opus 4.7 in Google Cloud | |||
| Claude Opus 4.6 in Google Cloud | |||
| Claude Opus 4.5 in Google Cloud | |||
| Claude Sonnet 4.6 in Google Cloud | |||
| Claude Sonnet 4.5 in Google Cloud | |||
| Claude Opus 4.1 in Google Cloud | |||
| Claude Haiku 4.5 in Google Cloud | |||
| Claude Opus 4 in Google Cloud | |||
| Claude Sonnet 4 in Google Cloud | |||
| Claude 3.7 Sonnet von Anthropic in Google Cloud | |||
| Anthropic Claude 3.5 Haiku in Google Cloud | |||
| Claude 3 Haiku von Anthropic in Google Cloud (eingestellt) | |||
| Mistral-Modelle | |||
| Mistral Medium 3 | |||
| Mistral OCR (25.05) | |||
| Mistral Small 3.1 (25.03) | |||
| Mistral Large (24.07) | |||
| Codestral 2 | |||
| Codestral (24.05) | |||
| Grok-Modelle | |||
| Grok 4.6 | |||
| Grok 4.20 (Reasoning) | |||
| Grok 4.20 (ohne Begründung) | |||
| Grok 4.1 Fast (Reasoning) | |||
| Grok 4.1 Fast (ohne Reasoning) | |||
Google Cloud Standorte für offene Modellendpunkte
Google verarbeitet Anfragen aus der von Ihnen angegebenen Region. Für einige Modelle bietet Google auch einen globalen Endpunkt an, um die allgemeine Verfügbarkeit zu verbessern und die Fehlerraten zu senken. Der globale Endpunkt kann ein separates Kontingent als der regionale Endpunkt haben und unterstützt keine Anforderungen an den Datenstandort. Weitere Informationen finden Sie unter Offene Modelle der Agent Platform für MaaS.
Offene Modellendpunkte für die Agent Platform sind in den folgenden Regionen verfügbar:
Global
| Global(global) | |
|---|---|
| Deepseek-Modelle | |
DeepSeek-OCR
(deepseek-ocr-maas)
|
|
DeepSeek-V3.2
(deepseek-v3.2-maas)
|
|
DeepSeek-V3.1
(deepseek-v3.1-maas)
|
|
DeepSeek R1 (0528)
(deepseek-r1-0528-maas)
|
|
| Google-Modelle | |
Gemma 4 26B A4B IT
experimental
(gemma-4-26b-a4b-it-maas)
|
|
| ZAI.org-Modelle | |
GLM 4.7
(glm-4.7-maas)
|
|
GLM 5
(glm-5-maas)
|
|
GLM 5.2
Vorschau
(glm-5.2-maas)
|
|
| OpenAI-Modelle | |
gpt-oss 120B
(gpt-oss-120b-maas)
|
|
gpt-oss 20B
(gpt-oss-20b-maas)
|
|
| Moonshot KI-Modelle | |
Kimi K2 Thinking
(kimi-k2-thinking-maas)
|
|
| Llama-Modelle | |
| Llama 3.3 70B | |
| Llama 4 Maverick 17B-128E | |
| Llama 4 Scout 17B-16E | |
| MiniMax-Modelle | |
MiniMax M2
(minimax-m2-maas)
|
|
| Qwen-Modelle | |
Qwen3-Next-80B Thinking
(qwen3-next-80b-a3b-thinking-maas)
|
|
Qwen3-Next-80B Instruct
(qwen3-next-80b-a3b-instruct-maas)
|
|
Qwen3 Coder
(qwen3-coder-480b-a35b-instruct-maas)
|
|
Qwen3 235B
(qwen3-235b-a22b-instruct-2507-maas)
|
|
| e5-Modelle | |
Mehrsprachig – E5 Small
(multilingual-e5-small-maas)
|
|
Multilingual E5 Large
(multilingual-e5-large-instruct-maas)
|
|
Multiregional
| USA (mehrere Regionen)(us) | EU (mehrere Regionen)(eu) | |
|---|---|---|
| Deepseek-Modelle | ||
DeepSeek-OCR
(deepseek-ocr-maas)
|
||
DeepSeek-V3.2
(deepseek-v3.2-maas)
|
||
DeepSeek-V3.1
(deepseek-v3.1-maas)
|
||
DeepSeek R1 (0528)
(deepseek-r1-0528-maas)
|
||
| Google-Modelle | ||
Gemma 4 26B A4B IT
experimental
(gemma-4-26b-a4b-it-maas)
|
||
| ZAI.org-Modelle | ||
GLM 4.7
(glm-4.7-maas)
|
||
GLM 5
(glm-5-maas)
|
||
GLM 5.2
Vorschau
(glm-5.2-maas)
|
||
| OpenAI-Modelle | ||
gpt-oss 120B
(gpt-oss-120b-maas)
|
||
gpt-oss 20B
(gpt-oss-20b-maas)
|
||
| Moonshot KI-Modelle | ||
Kimi K2 Thinking
(kimi-k2-thinking-maas)
|
||
| Llama-Modelle | ||
| Llama 3.3 70B | ||
| Llama 4 Maverick 17B-128E | ||
| Llama 4 Scout 17B-16E | ||
| MiniMax-Modelle | ||
MiniMax M2
(minimax-m2-maas)
|
||
| Qwen-Modelle | ||
Qwen3-Next-80B Thinking
(qwen3-next-80b-a3b-thinking-maas)
|
||
Qwen3-Next-80B Instruct
(qwen3-next-80b-a3b-instruct-maas)
|
||
Qwen3 Coder
(qwen3-coder-480b-a35b-instruct-maas)
|
||
Qwen3 235B
(qwen3-235b-a22b-instruct-2507-maas)
|
||
| e5-Modelle | ||
Mehrsprachig – E5 Small
(multilingual-e5-small-maas)
|
||
Multilingual E5 Large
(multilingual-e5-large-instruct-maas)
|
||
USA
| Oregon(us-west1) | Las Vegas(us-west4) | Iowa(us-central1) | South Carolina(us-east1) | N. Virginia(us-east4) | Columbus(us-east5) | Dallas(us-south1) | |
|---|---|---|---|---|---|---|---|
| Deepseek-Modelle | |||||||
DeepSeek-OCR
(deepseek-ocr-maas)
|
|||||||
DeepSeek-V3.2
(deepseek-v3.2-maas)
|
|||||||
DeepSeek-V3.1
(deepseek-v3.1-maas)
|
|||||||
DeepSeek R1 (0528)
(deepseek-r1-0528-maas)
|
|||||||
| Google-Modelle | |||||||
Gemma 4 26B A4B IT
experimental
(gemma-4-26b-a4b-it-maas)
|
|||||||
| ZAI.org-Modelle | |||||||
GLM 4.7
(glm-4.7-maas)
|
|||||||
GLM 5
(glm-5-maas)
|
|||||||
GLM 5.2
Vorschau
(glm-5.2-maas)
|
|||||||
| OpenAI-Modelle | |||||||
gpt-oss 120B
(gpt-oss-120b-maas)
|
|||||||
gpt-oss 20B
(gpt-oss-20b-maas)
|
|||||||
| Moonshot KI-Modelle | |||||||
Kimi K2 Thinking
(kimi-k2-thinking-maas)
|
|||||||
| Llama-Modelle | |||||||
| Llama 3.3 70B | |||||||
| Llama 4 Maverick 17B-128E | |||||||
| Llama 4 Scout 17B-16E | |||||||
| MiniMax-Modelle | |||||||
MiniMax M2
(minimax-m2-maas)
|
|||||||
| Qwen-Modelle | |||||||
Qwen3-Next-80B Thinking
(qwen3-next-80b-a3b-thinking-maas)
|
|||||||
Qwen3-Next-80B Instruct
(qwen3-next-80b-a3b-instruct-maas)
|
|||||||
Qwen3 Coder
(qwen3-coder-480b-a35b-instruct-maas)
|
|||||||
Qwen3 235B
(qwen3-235b-a22b-instruct-2507-maas)
|
|||||||
| e5-Modelle | |||||||
Multilingual E5 Small
(multilingual-e5-small-maas)
|
|||||||
Multilingual E5 Large
(multilingual-e5-large-instruct-maas)
|
|||||||
Nord- und Südamerika
| Montreal(northamerica-northeast1) | São Paulo(southamerica-east1) | |
|---|---|---|
| Deepseek-Modelle | ||
DeepSeek-OCR
(deepseek-ocr-maas)
|
||
DeepSeek-V3.2
(deepseek-v3.2-maas)
|
||
DeepSeek-V3.1
(deepseek-v3.1-maas)
|
||
DeepSeek R1 (0528)
(deepseek-r1-0528-maas)
|
||
| Google-Modelle | ||
Gemma 4 26B A4B IT
experimental
(gemma-4-26b-a4b-it-maas)
|
||
| ZAI.org-Modelle | ||
GLM 4.7
(glm-4.7-maas)
|
||
GLM 5
(glm-5-maas)
|
||
GLM 5.2
Vorschau
(glm-5.2-maas)
|
||
| OpenAI-Modelle | ||
gpt-oss 120B
(gpt-oss-120b-maas)
|
||
gpt-oss 20B
(gpt-oss-20b-maas)
|
||
| Moonshot KI-Modelle | ||
Kimi K2 Thinking
(kimi-k2-thinking-maas)
|
||
| Llama-Modelle | ||
| Llama 3.3 70B | ||
| Llama 4 Maverick 17B-128E | ||
| Llama 4 Scout 17B-16E | ||
| MiniMax-Modelle | ||
MiniMax M2
(minimax-m2-maas)
|
||
| Qwen-Modelle | ||
Qwen3-Next-80B Thinking
(qwen3-next-80b-a3b-thinking-maas)
|
||
Qwen3-Next-80B Instruct
(qwen3-next-80b-a3b-instruct-maas)
|
||
Qwen3 Coder
(qwen3-coder-480b-a35b-instruct-maas)
|
||
Qwen3 235B
(qwen3-235b-a22b-instruct-2507-maas)
|
||
| e5-Modelle | ||
Mehrsprachig – E5 Small
(multilingual-e5-small-maas)
|
||
Multilingual E5 Large
(multilingual-e5-large-instruct-maas)
|
||
Europa
| London(europe-west2) | Belgien(europe-west1) | Niederlande(europe-west4) | Zürich(europe-west6) | Frankfurt(europe-west3) | Finnland(europe-north1) | Warschau(europe-central2) | Mailand(europe-west8) | Madrid(europe-southwest1) | Paris(europe-west9) | |
|---|---|---|---|---|---|---|---|---|---|---|
| Deepseek-Modelle | ||||||||||
DeepSeek-OCR
(deepseek-ocr-maas)
|
||||||||||
DeepSeek-V3.2
(deepseek-v3.2-maas)
|
||||||||||
DeepSeek-V3.1
(deepseek-v3.1-maas)
|
||||||||||
DeepSeek R1 (0528)
(deepseek-r1-0528-maas)
|
||||||||||
| Google-Modelle | ||||||||||
Gemma 4 26B A4B IT
experimental
(gemma-4-26b-a4b-it-maas)
|
||||||||||
| ZAI.org-Modelle | ||||||||||
GLM 4.7
(glm-4.7-maas)
|
||||||||||
GLM 5
(glm-5-maas)
|
||||||||||
GLM 5.2
Vorschau
(glm-5.2-maas)
|
||||||||||
| OpenAI-Modelle | ||||||||||
gpt-oss 120B
(gpt-oss-120b-maas)
|
||||||||||
gpt-oss 20B
(gpt-oss-20b-maas)
|
||||||||||
| Moonshot KI-Modelle | ||||||||||
Kimi K2 Thinking
(kimi-k2-thinking-maas)
|
||||||||||
| Llama-Modelle | ||||||||||
| Llama 3.3 70B | ||||||||||
| Llama 4 Maverick 17B-128E | ||||||||||
| Llama 4 Scout 17B-16E | ||||||||||
| MiniMax-Modelle | ||||||||||
MiniMax M2
(minimax-m2-maas)
|
||||||||||
| Qwen-Modelle | ||||||||||
Qwen3-Next-80B Thinking
(qwen3-next-80b-a3b-thinking-maas)
|
||||||||||
Qwen3-Next-80B Instruct
(qwen3-next-80b-a3b-instruct-maas)
|
||||||||||
Qwen3 Coder
(qwen3-coder-480b-a35b-instruct-maas)
|
||||||||||
Qwen3 235B
(qwen3-235b-a22b-instruct-2507-maas)
|
||||||||||
| e5-Modelle | ||||||||||
Multilingual E5 Small
(multilingual-e5-small-maas)
|
||||||||||
Multilingual E5 Large
(multilingual-e5-large-instruct-maas)
|
||||||||||
Asiatisch-pazifischer Raum
| Mumbai(asia-south1) | Singapur(asia-southeast1) | Hongkong(asia-east2) | Taiwan(asia-east1) | Tokio(asia-northeast1) | Sydney(australia-southeast1) | Seoul(asia-northeast3) | |
|---|---|---|---|---|---|---|---|
| Deepseek-Modelle | |||||||
DeepSeek-OCR
(deepseek-ocr-maas)
|
|||||||
DeepSeek-V3.2
(deepseek-v3.2-maas)
|
|||||||
DeepSeek-V3.1
(deepseek-v3.1-maas)
|
|||||||
DeepSeek R1 (0528)
(deepseek-r1-0528-maas)
|
|||||||
| Google-Modelle | |||||||
Gemma 4 26B A4B IT
experimental
(gemma-4-26b-a4b-it-maas)
|
|||||||
| ZAI.org-Modelle | |||||||
GLM 4.7
(glm-4.7-maas)
|
|||||||
GLM 5
(glm-5-maas)
|
|||||||
GLM 5.2
Vorschau
(glm-5.2-maas)
|
|||||||
| OpenAI-Modelle | |||||||
gpt-oss 120B
(gpt-oss-120b-maas)
|
|||||||
gpt-oss 20B
(gpt-oss-20b-maas)
|
|||||||
| Moonshot KI-Modelle | |||||||
Kimi K2 Thinking
(kimi-k2-thinking-maas)
|
|||||||
| Llama-Modelle | |||||||
| Llama 3.3 70B | |||||||
| Llama 4 Maverick 17B-128E | |||||||
| Llama 4 Scout 17B-16E | |||||||
| MiniMax-Modelle | |||||||
MiniMax M2
(minimax-m2-maas)
|
|||||||
| Qwen-Modelle | |||||||
Qwen3-Next-80B Thinking
(qwen3-next-80b-a3b-thinking-maas)
|
|||||||
Qwen3-Next-80B Instruct
(qwen3-next-80b-a3b-instruct-maas)
|
|||||||
Qwen3 Coder
(qwen3-coder-480b-a35b-instruct-maas)
|
|||||||
Qwen3 235B
(qwen3-235b-a22b-instruct-2507-maas)
|
|||||||
| e5-Modelle | |||||||
Mehrsprachig – E5 Small
(multilingual-e5-small-maas)
|
|||||||
Multilingual E5 Large
(multilingual-e5-large-instruct-maas)
|
|||||||
Naher Osten
| Tel Aviv(me-west1) | Doha(me-central1) | Dammam(me-central2) | |
|---|---|---|---|
| Deepseek-Modelle | |||
DeepSeek-OCR
(deepseek-ocr-maas)
|
|||
DeepSeek-V3.2
(deepseek-v3.2-maas)
|
|||
DeepSeek-V3.1
(deepseek-v3.1-maas)
|
|||
DeepSeek R1 (0528)
(deepseek-r1-0528-maas)
|
|||
| Google-Modelle | |||
Gemma 4 26B A4B IT
experimental
(gemma-4-26b-a4b-it-maas)
|
|||
| ZAI.org-Modelle | |||
GLM 4.7
(glm-4.7-maas)
|
|||
GLM 5
(glm-5-maas)
|
|||
GLM 5.2
Vorschau
(glm-5.2-maas)
|
|||
| OpenAI-Modelle | |||
gpt-oss 120B
(gpt-oss-120b-maas)
|
|||
gpt-oss 20B
(gpt-oss-20b-maas)
|
|||
| Moonshot KI-Modelle | |||
Kimi K2 Thinking
(kimi-k2-thinking-maas)
|
|||
| Llama-Modelle | |||
| Llama 3.3 70B | |||
| Llama 4 Maverick 17B-128E | |||
| Llama 4 Scout 17B-16E | |||
| MiniMax-Modelle | |||
MiniMax M2
(minimax-m2-maas)
|
|||
| Qwen-Modelle | |||
Qwen3-Next-80B Thinking
(qwen3-next-80b-a3b-thinking-maas)
|
|||
Qwen3-Next-80B Instruct
(qwen3-next-80b-a3b-instruct-maas)
|
|||
Qwen3 Coder
(qwen3-coder-480b-a35b-instruct-maas)
|
|||
Qwen3 235B
(qwen3-235b-a22b-instruct-2507-maas)
|
|||
| e5-Modelle | |||
Mehrsprachig – E5 Small
(multilingual-e5-small-maas)
|
|||
Multilingual E5 Large
(multilingual-e5-large-instruct-maas)
|
|||