Ich brauche bei Openclaw hilfe undzwar bekomme ich immer diesen fehler ⚠️ Agent failed before reply: All models failed (4): xai/grok-4: Provider xai has auth issue (skipping all models) (auth) | xai/grok-3-fast-latest: Provider xai has auth issue (skipping all models) (auth) | xai/grok-3-mini: Provider xai has auth issue (skipping all models) (auth) | x-ai/grok-3-fast: No API key found for provider "x-ai". Auth store: /home/schule/.openclaw/agents/main/agent/auth-profiles.json (agentDir: /home/schule/.openclaw/agents/main/agent). Configure auth for this agent (openclaw agents add <id>) or copy auth-profiles.json from the main agentDir. (auth) | No API key found for provider "x-ai". Auth store: /home/schule/.openclaw/agents/main/agent/auth-profiles.json (agentDir: /home/schule/.openclaw/agents/main/agent). Configure auth for this agent (openclaw agents add <id>) or copy auth-profiles.json from the main agentDir. | No API key found for provider "x-ai". Auth store: /home/schule/.openclaw/agents/main/agent/auth-profiles.json (agentDir: /home/schule/.openclaw/agents/main/agent). Configure auth for this agent (openclaw agents add <id>) or copy auth-profiles.json from the main agentDir.
Logs: openclaw logs --follow
#Help my at discord/Openclaw
1 messages · Page 1 of 1 (latest)
Das ist ziemlich sicher kein Discord-Problem, sondern Model/Auth-Konfiguration.
Bitte zuerst posten:
openclaw --version
openclaw models status --probe
Wichtig: In OpenClaw heißt der native xAI/Grok Provider xai, nicht x-ai.
Wenn du direkt xAI nutzt
openclaw onboard --auth-choice xai-api-key
openclaw models set xai/grok-4
openclaw gateway restart
openclaw models status --probe
Docs: https://docs.openclaw.ai/providers/xai
Wenn du Grok über OpenRouter nutzt
Dann ist x-ai/... nur der OpenRouter-Modellpfad. Die OpenClaw-Model-ID muss so aussehen:
openclaw onboard --auth-choice openrouter-api-key
openclaw models set openrouter/x-ai/grok-3-fast
openclaw gateway restart
Docs: https://docs.openclaw.ai/providers/openrouter
Falls du den xAI-Key schon gesetzt hast: Auth ist pro Agent gespeichert unter ~/.openclaw/agents/<agentId>/agent/auth-profiles.json. Also unbedingt auf dem Gateway-Rechner und für den richtigen Agent konfigurieren. Bitte keine API Keys hier posten — nur redacted Output. 🦐
Config : ~/.openclaw/openclaw.json
Agent dir : ~/.openclaw/agents/main/agent
Default : xai/grok-4
Fallbacks (3) : xai/grok-4, xai/grok-3-fast-latest, xai/grok-3-mini
Image model : -
Image fallbacks (0): -
Aliases (1) : Grok -> xai/grok-4
Configured models (4): xai/grok-4, xai/grok-3-fast-latest, xai/grok-3-mini, x-ai/grok-3-fast
Auth overview
Auth store : ~/.openclaw/agents/main/agent/auth-profiles.json
Shell env : off
Providers w/ OAuth/tokens (0): -
- xai effective=profiles:~/.openclaw/agents/main/agent/auth-profiles.json | profiles=1 (oauth=0, token=0, api_key=1) | xai:default=xai-4f3M...y1DiVyQX
OAuth/token status
- none
Auth probes
┌──────────────────┬────────────────────────┬──────────────────────────────┐
│ Model │ Profile │ Status │
├──────────────────┼────────────────────────┼──────────────────────────────┤
│ xai/grok-4 │ xai:default (api_key) │ auth · 122.3s │
│ │ │ ↳ 403 "Your newly created │
│ │ │ team doesn't have any │
│ │ │ credits or licenses yet. │
│ │ │ You can purchase those on │
│ │ │ https://console.x.ai/team/ │
│ │ │ 91db36ad-c9d0-43c7-9307- │
│ │ │ dcdc34937893." │
└──────────────────┴────────────────────────┴──────────────────────────────┘
Probed 1 target in 204s
Der Probe-Output zeigt den eigentlichen Fehler klar:
Dein xAI API-Key ist da und wird gefunden, aber xAI lehnt ihn ab:
403 "Your newly created team doesn't have any credits or licenses yet"
Also: OpenClaw ist hier wahrscheinlich korrekt konfiguriert — dein xAI-Team hat nur noch keine Credits/Lizenz.
Mach das:
- In der xAI Console beim Team Credits/Lizenz hinzufügen:
https://console.x.ai/ - Danach kurz warten oder Gateway neu starten:
openclaw gateway restart openclaw models status --probe
Nebenbei: Entferne/ändere noch dieses Modell:
x-ai/grok-3-fast
Für direkten xAI-Zugriff muss es xai/... heißen. x-ai/... ist nur als OpenRouter-Pfad sinnvoll, dann aber vollständig:
openrouter/x-ai/grok-3-fast
Kurzfassung: Credits bei xAI kaufen/aktivieren, dann sollte xai/grok-4 funktionieren. 🦐
Gibt es kostenlose KIS?
Ja, aber mit Einschränkungen:
- Lokale KI über Ollama/LM Studio ist theoretisch kostenlos, aber für OpenClaw-Agenten oft schwach/langsam, außer du hast sehr starke Hardware.
- Kostenlose Provider-/Trial-Modelle gibt es manchmal, aber die ändern sich ständig und haben oft harte Limits.
- Für zuverlässige OpenClaw-Nutzung ist meistens ein bezahlter API-Key mit ein paar Credits deutlich weniger frustrierend.
Für aktuelle kostenlose/ günstige Modell-Tipps frag am besten in #models — die Community testet das laufend.
Docs:
https://docs.openclaw.ai/concepts/models
https://docs.openclaw.ai/concepts/models#model-discussions-community-notes
Ja aber ich will und habe kein Geld
Verständlich. Dann ist xAI/Grok direkt raus — ohne Credits/Lizenz funktioniert der API-Key nicht. Da gibt’s keinen kostenlosen Trick.
Die realistische kostenlose Option ist lokal über Ollama:
openclaw configure
Dann Model/auth → Ollama → Local only auswählen.
Oder grob manuell:
ollama serve
ollama pull <modellname>
openclaw models list --provider ollama
openclaw models set ollama/<modellname>
openclaw gateway restart
Aber ehrlich: kostenlose lokale Modelle können bei OpenClaw-Agenten langsam oder unzuverlässig sein, besonders ohne gute GPU/RAM.
Für aktuelle kostenlose Modell-Empfehlungen frag bitte in #models, weil sich Free-Tiers ständig ändern. Docs: https://docs.openclaw.ai/providers/ollama
Danke für die Hilfe ich werde zu Hause mal schauen
Gern! 🦐
Wenn du später weitermachst, am besten zuerst:
openclaw --version
openclaw models status --probe
Und keine API-Keys posten — nur gekürzte/zensierte Ausgaben. Wenn’s mit Ollama/local hakt, einfach hier reinschreiben.
Ich brauche hilfe
Klar — wobei genau hängst du gerade?
Bitte poste zuerst diese Ausgaben:
openclaw --version
openclaw models status --probe
Keine API-Keys posten, nur zensierte Ausgabe.
Falls es noch derselbe Fehler ist:
403 ... team doesn't have credits= xAI braucht Credits/Lizenz- kostenlos geht eher über Ollama lokal:
openclaw configure
Dann Model/auth → Ollama → Local only auswählen.
🦞 OpenClaw 2026.4.15 (041266a)
I'll butter your workflow like a lobster roll: messy, delicious, effective.
openclaw tui - ws://127.0.0.1:18789 - agent main - session main
history failed: GatewayClientRequestError: chat.history unavailable during
gateway startup
gateway connected | idle
agent main | session main | ollama/glm-5.1:cloud | tokens ?/203k
────────────────────────────────────────────────────────────────────────────
───────────
Das sieht erstmal nicht kaputt aus — der TUI ist verbunden:
gateway connected | idle
agent main | session main | ollama/glm-5.1:cloud
Die Zeile hier ist oft nur beim Start kurzzeitig:
history failed ... chat.history unavailable during gateway startup
Wenn der Gateway gerade hochfährt: 10–20 Sekunden warten, TUI schließen/neu öffnen.
Wichtiger Punkt: Du nutzt gerade:
ollama/glm-5.1:cloud
Das ist Ollama Cloud, nicht lokal. Wenn du wirklich kostenlos/lokal willst, brauchst du ein lokales Modell ohne :cloud, z.B.:
ollama serve
ollama pull gemma4
openclaw models list --provider ollama
openclaw models set ollama/gemma4
openclaw gateway restart
Falls danach Antworten fehlschlagen, poste bitte:
openclaw models status --probe
openclaw logs --tail 80
OpenClaw: access not configured.
Your Discord user id: 1018236519204532345
Pairing code:
2ZEKLGDC
Ask the bot owner to approve with:
openclaw pairing approve discord 2ZEKLGDC
openclaw pairing approve discord 2ZEKLGDC
Das ist jetzt ein Pairing-/Zugriffsproblem, nicht mehr das Modell.
Auf dem Rechner, wo dein Gateway läuft, ausführen:
openclaw pairing approve discord 2ZEKLGDC
Danach deine Discord-Nachricht nochmal senden/testen.
Wenn es nicht klappt:
openclaw pairing list discord
openclaw logs --tail 80
Wichtig: Nur du/der Bot-Owner sollte Pairing-Codes genehmigen. In deinem Fall passt die User-ID aus der Meldung zu dir, also ist das genau der richtige Schritt.
Ich kann aber in der Console nicht schreiben
Du bist vermutlich gerade im OpenClaw TUI — dort kannst du keine normalen Shell-Befehle wie openclaw pairing approve ... eingeben.
Mach eins davon:
Option A: Neues Terminal / neue Konsole öffnen und dort ausführen:
openclaw pairing approve discord 2ZEKLGDC
Option B: Den TUI beenden mit Ctrl+C, dann im normalen Terminal ausführen:
openclaw pairing approve discord 2ZEKLGDC
Danach Discord nochmal testen.
Wie kann ich Grok rausschmeißen?
Ja. In einem normalen Terminal — nicht im TUI — zuerst ein anderes Modell setzen, dann Grok aus den Fallbacks/Alias rausnehmen.
Wenn du bei Ollama bleiben willst, z.B.:
openclaw models set ollama/glm-5.1:cloud
openclaw models fallbacks clear
openclaw models aliases remove Grok
openclaw gateway restart
openclaw models status
Wenn du Grok auch aus der „Configured models“-Liste entfernen willst, ersetze die Allowlist komplett mit deinem gewünschten Modell:
openclaw config set agents.defaults.models '{"ollama/glm-5.1:cloud":{}}' --strict-json --replace
openclaw gateway restart
openclaw models status
Falls du lokal/kostenlos willst, nimm statt ollama/glm-5.1:cloud ein lokales Modell ohne :cloud, z.B. nach ollama pull gemma4:
openclaw models set ollama/gemma4
openclaw config set agents.defaults.models '{"ollama/gemma4":{}}' --strict-json --replace
Wird mein Discord Bot dann eigentlich 24/7 von dir gehostet?
Nein — ich hoste deinen Bot nicht.
OpenClaw läuft auf deinem Rechner/VPS. Dein Discord-Bot ist nur online, solange dein OpenClaw Gateway läuft und Internet hat.
Für 24/7 brauchst du z.B.:
- deinen PC dauerhaft anlassen
- oder einen VPS / Server / Raspberry Pi
- und OpenClaw als Service laufen lassen
Prüfen kannst du es mit:
openclaw gateway status
Starten/neu starten:
openclaw gateway start
openclaw gateway restart
Wenn dein Rechner aus ist, ist der Bot auch weg.
404 {"error":"model 'gemma4' not found"} kommt in discord
404 model 'gemma4' not found heißt: Der Ollama-Server, den OpenClaw benutzt, hat dieses Modell nicht installiert/gefunden.
Mach im normalen Terminal:
ollama list
Nimm dann exakt den Namen aus der Liste:
openclaw models set ollama/<EXAKTER_NAME_AUS_OLLAMA_LIST>
openclaw gateway restart
Wenn kein Modell installiert ist: