-
Notifications
You must be signed in to change notification settings - Fork 2
Expand file tree
/
Copy path.env.example
More file actions
59 lines (47 loc) · 2.09 KB
/
Copy path.env.example
File metadata and controls
59 lines (47 loc) · 2.09 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
# Inference configuration for the agent-shaped examples.
#
# Copy to .env, uncomment exactly ONE block, and load it into your shell:
#
# cp .env.example .env
# set -a; . ./.env; set +a
#
# Nothing here is active by default and the examples have no built-in
# fallback: with none of these set they exit immediately with a message
# naming what to export.
#
# The examples never special-case a provider. INTX_BASE_URL, INTX_MODEL
# and INTX_API_KEY are handed straight to Interchange's inference layer;
# INTX_PROVIDER only picks the wire format (anthropic, openai,
# openai-compatible, google-genai) and defaults to openai-compatible.
# --- Local Ollama (https://ollama.com) ---------------------------------
# Serves an OpenAI-compatible API. It ignores the Authorization header,
# but a source always carries a credential, so put a placeholder in.
# `ollama list` shows the models you have pulled.
#
# INTX_BASE_URL=http://localhost:11434/v1
# INTX_MODEL=qwen2.5vl:7b
# INTX_API_KEY=ollama
# --- Anthropic ---------------------------------------------------------
# ANTHROPIC_API_KEY alone implies the endpoint and the adapter; INTX_MODEL
# is still required, because model ids change faster than this file does.
#
# ANTHROPIC_API_KEY=sk-ant-...
# INTX_MODEL=claude-sonnet-4-6
# --- OpenAI ------------------------------------------------------------
# OPENAI_API_KEY=sk-...
# INTX_MODEL=gpt-4o-mini
# --- Google Gemini -----------------------------------------------------
# GOOGLE_API_KEY=...
# INTX_MODEL=gemini-2.0-flash
# --- Any other OpenAI-compatible endpoint ------------------------------
# vLLM, LM Studio, llama.cpp, OpenRouter, Together, a gateway of your own.
# Same three variables, no code change.
#
# INTX_BASE_URL=https://openrouter.ai/api/v1
# INTX_MODEL=meta-llama/llama-3.3-70b-instruct
# INTX_API_KEY=sk-or-...
# --- Postgres (modules/ and solutions/inbox-triage) --------------------
# Defaults to postgres://postgres:postgres@localhost:5459/ex_<name>.
# See modules/README.md for a docker one-liner that starts one.
#
# EXAMPLES_DATABASE_URL=postgres://postgres:postgres@localhost:5459/ex_mailbox