Skip to main content
Find answers to the most common questions about setting up and running AI strategies on Everstrike. If you’re new to strategies, start with Strategies and Credits for full context.

Setup & Risk

Start at a 90% minimum confidence. Lower thresholds — such as 85% or 80% — can produce more frequent trades but may also increase the rate of incorrect decisions.See Strategies for a breakdown of common confidence settings and their trade-offs.
Start with a low order quantity, such as $10, or test your strategy on Everstrike Testnet first. Validate that the strategy’s events look correct before increasing the quantity.
Choose a timeframe that matches your intended trading horizon. Lower timeframes generally produce more frequent trades and shorter holding periods; higher timeframes produce fewer, longer-duration positions.
Add a maximum position size directly to your strategy prompt. The AI respects position limits you provide.For example:
Buy the dip on BTC. Maximum position size: $3,000.
Also start with a low order quantity while you validate the strategy’s behavior on live markets before scaling up.
Yes. Describe the conditions for closing the position in your prompt and the AI will act on them.For example:
Buy the dip on BTC. Close the position when the dip is over.
You can also instruct the AI to maintain a specific target position size, automatically reducing or increasing the position as the market moves.

Strategy Limits & Behavior

You can run up to 10 strategies at the same time.
Each strategy can monitor up to 10 markets. With 10 active strategies running in parallel, the AI can watch up to 100 markets simultaneously.
No. The nickname is for display purposes only and for associating events with the correct strategy in the UI. It has no effect on how the AI evaluates markets or places orders.
Every strategy — regardless of the model or evaluation frequency you choose — produces its first event within 20 seconds of creation. If the event’s confidence score exceeds your minimum threshold, it can lead to a trade immediately.Track events in real time through the frontend or through the API.
If an active strategy misses an evaluation cycle, the model may not be processing your prompt quickly enough to keep up with the selected evaluation frequency. Try choosing a more powerful model or simplifying the strategy prompt to reduce processing time.
Yes. The AI can manage up to 10 limit orders per strategy, including for grid trading setups where orders are spread across price levels.

Prompts & AI Limits

See Supported Data for the full list of inputs available to your strategy. The AI cannot access data outside that list and cannot perform visual chart analysis on price charts.
You can write prompts in any commonly used natural language, including English, Spanish, Norwegian, and others. English generally produces the best results.
Yes. Specify your preferred language in the strategy prompt and the AI will produce its reasoning output in that language. The default reasoning language is English.
Use a minimum confidence of at least 90%, set temperature to 0, and disable discretion. With these settings, the hallucination rate is significantly below 0.1% — meaning the AI makes incorrect decisions on fewer than 1 in 1,000 evaluations.

Costs & Credits

The default GPT-4.1-mini model costs $0.00015 per event.At a one-second evaluation frequency, one market generates 3,600 events per hour — a cost of $0.54 per hour. Costs scale linearly with every additional market your strategy watches.See strategy costs for a full comparison across models and evaluation frequencies.
No. Paused strategies do not generate events, place trades, or consume any credits. Pause a strategy at any time to stop all credit consumption immediately.
Fund your account with at least $5 (or the equivalent in USDC or USDT). Once that threshold is met, strategy expenses are deducted directly from your USD balance.See Credits for details on free grants, daily replenishment, and funding options.

Frequency

No. The fastest available evaluation frequency is one second, which supports low- and mid-frequency strategies but not high-frequency trading.Future releases may add shorter evaluation frequencies such as 100 ms and 10 ms.