mirror of
https://github.com/trustgraph-ai/trustgraph.git
synced 2026-07-01 09:29:38 +02:00
72870a7e2e
Unit tests: Consumer class (7), recursive-splitter (10), parseJsonResponse (11) — 28 total. Docker: add 5 commented LLM provider services, dev compose override, .env.example. Workbench: chat persistence, error boundary, disconnect banner, prompts error handling. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
24 lines
495 B
Text
24 lines
495 B
Text
# LLM Provider API Keys (set the one matching your active text-completion service)
|
|
OPENAI_TOKEN=
|
|
OPENAI_BASE_URL=
|
|
CLAUDE_KEY=
|
|
AZURE_TOKEN=
|
|
AZURE_ENDPOINT=
|
|
AZURE_MODEL=gpt-4o
|
|
AZURE_API_VERSION=2024-02-15-preview
|
|
OLLAMA_MODEL=gemma3:4b
|
|
OPENAI_COMPAT_URL=http://localhost:1234/v1
|
|
OPENAI_COMPAT_MODEL=default
|
|
OPENAI_COMPAT_KEY=
|
|
MISTRAL_TOKEN=
|
|
MISTRAL_MODEL=ministral-8b-latest
|
|
|
|
# Gateway
|
|
GATEWAY_PORT=8088
|
|
GATEWAY_SECRET=
|
|
|
|
# Workbench
|
|
WORKBENCH_PORT=3001
|
|
|
|
# Grafana
|
|
GF_SECURITY_ADMIN_PASSWORD=admin
|