-
Notifications
You must be signed in to change notification settings - Fork 10
Expand file tree
/
Copy path.env.example
More file actions
202 lines (174 loc) · 14.1 KB
/
Copy path.env.example
File metadata and controls
202 lines (174 loc) · 14.1 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
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
# ╔══════════════════════════════════════════════════════════════════════════╗
# ║ QUORUM CONFIGURATION ║
# ╚══════════════════════════════════════════════════════════════════════════╝
#
# Copy this file to .env and fill in your API keys and preferred models.
#
# IMPORTANT: Cloud provider models (OpenAI, Anthropic, Google, xAI) must be
# listed in their respective *_MODELS variables to be available in Quorum.
# Ollama models are auto-discovered and don't need to be listed.
# ══════════════════════════════════════════════════════════════════════════════
# OPENAI
# ══════════════════════════════════════════════════════════════════════════════
# Get your API key: https://platform.openai.com/api-keys
# Available models: https://platform.openai.com/docs/models
#
OPENAI_API_KEY=sk-...
OPENAI_MODELS=gpt-5.2,gpt-5.1,gpt-5
# ══════════════════════════════════════════════════════════════════════════════
# ANTHROPIC
# ══════════════════════════════════════════════════════════════════════════════
# Get your API key: https://console.anthropic.com/settings/keys
# Available models: https://docs.anthropic.com/en/docs/about-claude/models
#
ANTHROPIC_API_KEY=sk-ant-...
ANTHROPIC_MODELS=claude-opus-4-5-20251124,claude-sonnet-4-5-20250929
# ══════════════════════════════════════════════════════════════════════════════
# GOOGLE
# ══════════════════════════════════════════════════════════════════════════════
# Get your API key: https://aistudio.google.com/apikey
# Available models: https://ai.google.dev/gemini-api/docs/models
#
GOOGLE_API_KEY=...
GOOGLE_MODELS=gemini-3-pro,gemini-2.5-flash
# ══════════════════════════════════════════════════════════════════════════════
# XAI (GROK)
# ══════════════════════════════════════════════════════════════════════════════
# Get your API key: https://console.x.ai/
# Available models: https://docs.x.ai/docs/models
#
XAI_API_KEY=xai-...
XAI_MODELS=grok-4.1,grok-4
# ══════════════════════════════════════════════════════════════════════════════
# OLLAMA (LOCAL MODELS)
# ══════════════════════════════════════════════════════════════════════════════
# Install Ollama: https://ollama.com/download
# Available models: https://ollama.com/library
#
# Models are auto-discovered - no configuration needed for same-machine setup!
# Just run "ollama pull llama3" and it appears in /models automatically.
#
# Only configure these if Ollama runs on a different machine:
# OLLAMA_BASE_URL=http://localhost:11434
# OLLAMA_API_KEY= # Optional, for proxy authentication
#
# WSL users with Ollama on Windows: See README.md troubleshooting section
# ══════════════════════════════════════════════════════════════════════════════
# OPENAI-COMPATIBLE PROVIDERS
# ══════════════════════════════════════════════════════════════════════════════
# Connect to any service that uses the OpenAI API format.
# These work alongside native providers - you can use them simultaneously.
# ──────────────────────────────────────────────────────────────────────────────
# OpenRouter (Multi-model API aggregator)
# ──────────────────────────────────────────────────────────────────────────────
# Access Claude, GPT, Llama, Mistral and 200+ models through one API.
# Get your API key: https://openrouter.ai/keys
#
# OPENROUTER_API_KEY=sk-or-v1-...
# OPENROUTER_MODELS=anthropic/claude-3-opus,openai/gpt-4o,meta-llama/llama-3.1-70b
#
# Optional: Override base URL (default: https://openrouter.ai/api/v1)
# OPENROUTER_BASE_URL=https://openrouter.ai/api/v1
# ──────────────────────────────────────────────────────────────────────────────
# LM Studio (Local desktop app)
# ──────────────────────────────────────────────────────────────────────────────
# Run local models with a nice GUI. Download: https://lmstudio.ai/
# No API key required - just list your loaded models.
#
# LMSTUDIO_MODELS=llama-3.2-3b,deepseek-coder-v2
#
# Optional: Override base URL (default: http://localhost:1234/v1)
# LMSTUDIO_BASE_URL=http://localhost:1234/v1
# LMSTUDIO_API_KEY= # Optional, usually not needed
# ──────────────────────────────────────────────────────────────────────────────
# llama-swap (Hot-swap server for local models)
# ──────────────────────────────────────────────────────────────────────────────
# Automatically load/unload models based on requests.
# GitHub: https://github.com/mostlygeek/llama-swap
#
# LLAMASWAP_BASE_URL=http://localhost:8080/v1
# LLAMASWAP_MODELS=llama3,mistral-7b,qwen2
# LLAMASWAP_API_KEY= # Optional, usually not needed
# ──────────────────────────────────────────────────────────────────────────────
# Custom OpenAI-compatible endpoint
# ──────────────────────────────────────────────────────────────────────────────
# Generic fallback for any OpenAI-compatible server not listed above.
# Examples: vLLM, LocalAI, text-generation-webui, TabbyAPI, etc.
#
# CUSTOM_BASE_URL=http://localhost:5000/v1
# CUSTOM_MODELS=model-name-1,model-name-2
# CUSTOM_API_KEY= # Required if your server needs auth
# ══════════════════════════════════════════════════════════════════════════════
# QUORUM SETTINGS
# ══════════════════════════════════════════════════════════════════════════════
# Language setting
# Leave empty/unset to respond in the same language as the question
# Or set a specific language to always use
# QUORUM_DEFAULT_LANGUAGE=Swedish
# ──────────────────────────────────────────────────────────────────────────────
# Discussion Methods
# ──────────────────────────────────────────────────────────────────────────────
# Choose your method in the terminal UI. Each has authentic structure:
#
# STANDARD (2+ models) - Configurable consensus-seeking discussion
# OXFORD (even number) - Formal parliamentary debate with FOR/AGAINST teams
# ADVOCATE (3+ models) - Devil's advocate challenges the consensus
# SOCRATIC (2+ models) - Question-driven Socratic dialogue
# DELPHI (3+ models) - Iterative consensus for estimates and forecasts
# BRAINSTORM (2+ models) - Creative ideation (diverge → build → converge)
# TRADEOFF (2+ models) - Structured comparison of alternatives
#
# NOTE: Only Standard method uses the settings below.
# Other methods have fixed authentic structures.
# ──────────────────────────────────────────────────────────────────────────────
# Standard Method Settings
# ──────────────────────────────────────────────────────────────────────────────
# Rounds per agent in Standard's Phase 3 discussion (default: 2)
# Total discussion messages = rounds_per_agent * number_of_models
QUORUM_ROUNDS_PER_AGENT=2
# Which model creates the final synthesis in Standard's Phase 5
# first - Always use the first selected model (default, predictable)
# random - Randomly select a model each time
# rotate - Rotate through models across discussions
QUORUM_SYNTHESIZER=first
# ──────────────────────────────────────────────────────────────────────────────
# Model Timeout
# ──────────────────────────────────────────────────────────────────────────────
# Maximum time to wait for a model response (in seconds).
# Default: 60 seconds
# Increase for slow local models (e.g., Ollama on CPU): 120-300
#
# QUORUM_MODEL_TIMEOUT=60
# ──────────────────────────────────────────────────────────────────────────────
# Execution Mode (VRAM Optimization)
# ──────────────────────────────────────────────────────────────────────────────
# Controls how models are executed in parallel phases.
# This is critical for local Ollama models that share GPU VRAM.
#
# auto - Smart default: Cloud APIs run in parallel, Ollama runs sequentially
# parallel - Always run all models simultaneously (best for cloud-only setups)
# sequential - Always run models one at a time (best for limited VRAM)
#
# Default: auto (recommended for most users)
#
# QUORUM_EXECUTION_MODE=auto
# ──────────────────────────────────────────────────────────────────────────────
# Auto-save Reports
# ──────────────────────────────────────────────────────────────────────────────
# All discussions are automatically saved as markdown files.
# Configure the directory here (supports ~ for home directory expansion).
# Default: ~/reports
#
# QUORUM_REPORT_DIR=~/reports
# ──────────────────────────────────────────────────────────────────────────────
# Export Settings
# ──────────────────────────────────────────────────────────────────────────────
# Directory for exported discussions (manual /export command).
# Supports ~ for home directory expansion.
# If not set, exports to home directory (~).
#
# QUORUM_EXPORT_DIR=~/.quorum/exports
# Default export format: md (markdown), text (plain text), pdf, json
# Can be overridden per export: /export text, /export pdf, /export json
#
# QUORUM_EXPORT_FORMAT=md