-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy path.env.example
More file actions
98 lines (87 loc) · 5.05 KB
/
Copy path.env.example
File metadata and controls
98 lines (87 loc) · 5.05 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
# Environment Configuration for IWAC AI Pipelines
# Copy this file to .env and fill in your actual values
# =============================================================================
# OMEKA S API CONFIGURATION
# =============================================================================
# Base URL for your Omeka S instance API
# Format: https://your-omeka-instance.com/api
OMEKA_BASE_URL=https://your-omeka-instance.com/api
# API Key Identity (generated in Omeka S admin panel)
OMEKA_KEY_IDENTITY=your_key_identity_here
# API Key Credential (generated in Omeka S admin panel)
OMEKA_KEY_CREDENTIAL=your_key_credential_here
# =============================================================================
# GOOGLE GEMINI API CONFIGURATION
# =============================================================================
# Google Gemini API Key (get from Google AI Studio)
# URL: https://aistudio.google.com/app/apikey
GEMINI_API_KEY=your_gemini_api_key_here
# =============================================================================
# OPENAI (CHATGPT) API CONFIGURATION
# =============================================================================
# OpenAI API Key (for ChatGPT-enabled pipelines)
# Get an API key from: https://platform.openai.com/api-keys
# Ensure your account has access to the GPT-5.6 family: gpt-5.6-luna (default,
# cost-optimized), gpt-5.6-terra (balanced), gpt-5.6-sol (flagship).
# Use the --model flag to pick the tier per run.
OPENAI_API_KEY=your_openai_api_key_here
# =============================================================================
# MISTRAL AI API CONFIGURATION
# =============================================================================
# Mistral AI API Key (for Mistral Large 3 and Document AI pipelines)
# Get an API key from: https://console.mistral.ai/home
# Used for mistral-large-2512 (text generation), mistral-ocr-latest (OCR processing)
MISTRAL_API_KEY=your_mistral_api_key_here
# =============================================================================
# OPENROUTER API CONFIGURATION
# =============================================================================
# OpenRouter API Key — one key for the open-weights models (Qwen, DeepSeek)
# Get an API key from: https://openrouter.ai/keys
# DeepSeek V4 Flash 0731 is the default for every text-generation stage.
# Multimodal extraction still uses Gemini, Mistral, or Voxtral, then hands
# extracted text to DeepSeek where a separate consolidation stage exists.
# Requests are pinned to backends that do not retain data (data_collection=deny).
OPENROUTER_API_KEY=your_openrouter_api_key_here
# Optional: set to 1 to route only to zero-data-retention backends (OpenRouter's
# `zdr` flag). Stricter than the default training opt-out; fewer providers
# qualify, and a model with none fails with a 503.
# OPENROUTER_ZDR=1
# =============================================================================
# SELF-HOSTED MODEL ENDPOINT (OPTIONAL)
# =============================================================================
# Any OpenAI-compatible endpoint you run yourself: vLLM on a GPU cluster,
# llama.cpp, LM Studio, TGI, Ollama. Leave both blank unless you have one —
# models on this route are simply reported as unavailable when the URL is unset.
#
# See serving/README.md for the full walkthrough, including a worked example on
# a Slurm cluster with an SSH tunnel. Reaching a cluster GPU node usually means
# forwarding a port to your machine, which is why this is a localhost address:
# ssh -L 8000:<compute-node>:8000 <user>@<login-host>
SELFHOSTED_LLM_BASE_URL=
# Must match the --api-key the server was started with. Leave blank for a server
# started without one. Generate with: openssl rand -hex 24
# Keep the sk- prefix: it is what the log redaction in common/log_redaction.py
# recognises, so an accidental echo into a log file stays masked.
SELFHOSTED_LLM_API_KEY=
# =============================================================================
# FFMPEG CONFIGURATION (for Audio Transcription)
# =============================================================================
# Path to ffmpeg executable (required for audio splitting in AI_audio_summary)
# Download from: https://ffmpeg.org/download.html or install via 'winget install Gyan.FFmpeg'
# Leave empty if ffmpeg is already on your system PATH
FFMPEG_PATH=
FFPROBE_PATH=
# =============================================================================
# SETUP INSTRUCTIONS
# =============================================================================
# 1. Copy this file to .env in the same directory
# 2. Fill in your actual API credentials
# 3. Ensure .env is in your .gitignore file
# 4. Test your configuration by running the scripts
# =============================================================================
# SECURITY NOTES
# =============================================================================
# - Never commit the .env file to version control
# - Keep your API keys secure and rotate them regularly
# - Use environment-specific configurations for different deployments
# - Consider using a secrets management service for production