Configuration

Complete guide to configuring Gopher for optimal performance.

Settings Overview

Gopher settings are organized into four sections:

Section	Contents
General	Data source, API keys, authentication
Models	LLM model selection
Trading	Asset, timeframes, strategy preferences
Advanced	Iteration limits, context settings

General Settings

Data Source

Option	Description
Cloud API	Uses hosted data service. No setup required. (Default)
Local Database	Connect to your own PostgreSQL instance.

The cloud API is recommended for most users - it provides access to historical data for 200+ assets without any setup.

Authentication

Gopher supports multiple ways to access LLM inference. Configure at least one:

Gopher Credits (Recommended)

The easiest way to get started - use Gopher's hosted inference service:

1. Go to gotrader.gopher-ai.com/settings
2. Create an account or sign in
3. Purchase credits (or use existing balance)
4. Copy your Gopher Key (format: gopher_xxx...)
5. Paste in the Gopher Key field in Settings
6. Click Test to verify

Benefits of Gopher Credits:

• No external API keys needed
• Optimized models pre-configured
• Pay-as-you-go pricing
• Single account for everything

OpenRouter

Bring your own OpenRouter API key:

1. Get a key at openrouter.ai/keys
2. Paste in the OpenRouter API Key field
3. Click Test to verify

OpenRouter provides access to 100+ models with a single API key.

OpenAI

Use OpenAI directly:

1. Get a key at platform.openai.com/api-keys
2. Paste in the OpenAI API Key field
3. Click Test to verify

Inference

Gopher resolves inference settings per loop, which means the main evolution loop, tools, backtests, Monte Carlo validation, and compression can each use different providers or models when needed.

Inference providers:

• Gopher Credits
• OpenRouter
• OpenAI
• Ollama (local)
• Basilica (OpenAI-compatible)
• Custom (OpenAI-compatible)

Per-loop overrides: You can set overrides per loop (provider, model ID, base URL, API key) in the desktop app: Settings → Advanced → Inference Overrides.

Base URL conventions:

• Use the OpenAI-compatible root URL only (no /chat/completions).
• When using Gopher Credits, the base URL is normalized to .../api/v1/gopher.

Environment variables:

• BART_GOPHER_CODE
• BART_GOPHER_API_URL
• OPENROUTER_API_KEY
• OPENAI_API_KEY

Basilica (OpenAI-compatible)

Basilica deployments expose an OpenAI-compatible API. See:

• https://docs.basilica.ai/introduction
• https://docs.basilica.ai/inference

Setup steps:

1. Deploy a model with Basilica (vLLM or SGLang) and copy the deployment URL.
2. Configure Gopher with:
   • Base URL: ${DEPLOYMENT_URL}/v1
   • API Key: not-needed (Basilica handles auth for the deployment URL)
3. Example model IDs (Hugging Face IDs):
   • Qwen/Qwen2.5-7B-Instruct
   • meta-llama/Llama-3.1-8B-Instruct
   • mistralai/Mistral-7B-Instruct-v0.3
   • deepseek-ai/DeepSeek-V2-Chat

Model Settings

Loop Model

The primary model for strategy generation and evolution.

Recommended models:

Model	Provider	Notes
qwen/qwen3-max	OpenRouter	Best quality, recommended
anthropic/claude-3.5-sonnet	OpenRouter	Fast and capable
openai/gpt-4-turbo	OpenRouter/OpenAI	Reliable fallback

Backtest Model

A faster model used for trade decisions during backtests.

Recommended models:

Model	Provider	Notes
qwen/qwen3-vl-8b-instruct	OpenRouter	Fast and efficient
openai/gpt-4o-mini	OpenRouter/OpenAI	Good balance

Adding Custom Models

Click + Add Custom to add models not in the default list:

Field	Description
Model ID	The model identifier (e.g., deepseek-v3)
Display Name	Friendly name shown in dropdowns
Provider	OpenRouter, OpenAI, Ollama, or Custom
Base URL	API endpoint (auto-filled for known providers)
API Key	Key for this specific model (if required)

Trading Settings

Default Asset

Pre-selected asset for new evolutions. Popular choices:

• BTC - Bitcoin, most data coverage
• ETH - Ethereum
• SOL - Solana

Minimum Coverage (%)

Minimum data coverage required to run a backtest. Default: 65%.

Higher values ensure more reliable backtests but limit date range options.

Timeframe Intervals

Select intervals for multi-timeframe analysis:

Interval	Use Case
1m, 5m	Scalping
15m, 30m	Day trading
1h, 4h, 8h	Swing trading
1d, 1w	Position trading

Tip: Use multiple timeframes together (e.g., 15m + 1h + 4h) for better results.

Indicators

Indicators control how the AI summarizes market conditions across timeframes.

• Configure indicators per run in Backtest, Evolution, or Monte Carlo
• Saved presets include the indicator set
• If no indicators are selected, defaults are used

See the dedicated guide for examples and full configuration:

• Indicators

Default Strategy Prompt

Optional starting prompt for evolution. Examples:

• "Focus on trend-following strategies with clear entry/exit rules"
• "Use mean reversion with Bollinger Bands"
• Leave empty for fully autonomous discovery

Advanced Settings

Max Iterations per Asset

Number of strategy evolution cycles. Default: 50.

Value	Trade-off
10-25	Quick exploration, lower API costs
50	Balanced (recommended)
100+	Thorough exploration, higher API costs

Max Tool Calls per Iteration

Maximum backtests within one iteration. Default: 10.

Higher values allow the AI to explore more variations per iteration but increase API costs.

Evaluation Interval

How often to evaluate positions during backtests. Default: 15m.

Shorter intervals = more decisions = higher costs but finer granularity.

Monte Carlo Runs

Number of Monte Carlo simulations for validation. Default: 100.

Risk Settings

Configure risk parameters in the Evolution panel:

Setting	Default	Description
Risk %	2%	Percentage of equity risked per trade
Stop %	0.5%	Minimum stop-loss distance
Target R:R	2:1	Target reward-to-risk ratio
Min R:R	1:1	Minimum R:R enforced
Max Margin	25%	Maximum margin utilization
Leverage	10x	Position leverage
Initial Balance	$10,000	Starting capital for backtests

Environment Variables

For CLI users or advanced configuration:

Variable	Description
BART_GOPHER_CODE	Gopher Key for hosted inference (recommended)
OPENROUTER_API_KEY	OpenRouter API key
OPENAI_API_KEY	OpenAI API key
DATABASE_URL	Local database connection string

Example setup:

# ~/.bashrc or ~/.zshrc

# Option 1: Gopher Credits (recommended)
export BART_GOPHER_CODE='gopher_your-key-here'

# Option 2: OpenRouter
export OPENROUTER_API_KEY='sk-or-v1-your-key-here'

Get your Gopher Key at gotrader.gopher-ai.com/settings.

Presets

Save and load configuration presets for different strategies:

Saving a Preset

1. Configure your desired settings
2. Go to Evolution tab
3. Click Save Preset
4. Enter a name (e.g., "BTC Swing Trading")

Loading a Preset

1. Open the preset dropdown in the Evolution tab
2. Select your saved preset
3. All settings are restored

Built-in Presets

Gopher includes several default presets:

Preset	Description
Quick Test	Fast iteration for testing (10 iterations)
Standard	Balanced settings (50 iterations)
Thorough	Deep exploration (100 iterations)
Scalping	Short timeframes (1m, 5m)
Swing	Medium timeframes (1h, 4h, 8h)
Position	Long timeframes (1d, 1w)

Configuration File Location

Settings are stored locally:

Platform	Location
macOS	~/Documents/Gopher/config.json
Windows	%USERPROFILE%\Documents\Gopher\config.json
Linux	~/Documents/Gopher/config.json

Next Steps

• Strategy Evolution - Learn how evolution works
• Data Coverage - Available assets and timeframes
• Supported Models - Full list of LLM options