Braiins Price Adapt early access ends September 30: what automated power targeting actually changes
Source report: 2026-08-20 · Editorial analysis published: 2026-09-28
Braiins says Price Adapt can adjust each supported miner to a profitable power target or curtail it when expected revenue no longer covers electricity. Free early access runs through September 30, while published gains are modelled results that operators should validate against their own tariff, fleet and pool data.
Analysis and practical implications
This section is our analysis and illustrative calculations, separate from the source report.
Price Adapt links miner settings to market conditions
Braiins describes Price Adapt as an automation layer inside Braiins Manager that evaluates forecast electricity prices, hashprice and the efficiency curve of each supported miner. It can raise a power target when energy is cheap, reduce the target when margins tighten, or stop hashing when expected revenue falls below power cost. The product is in free early access through September 30, 2026. That deadline makes the current evaluation practical, but it does not imply that every farm should enable automatic control without a staged test.
The published percentages are modelling results
Braiins reports a modelled profit improvement of 36.2% for an Antminer S19J Pro and 15.3% for an Antminer S21 XP in a 1 MW ERCOT example using 2025 day-ahead prices, a seven-day trailing hashprice and Braiins OS efficiency curves. The company separates the firmware contribution from the Price Adapt contribution. These figures are useful as a scenario, not a guaranteed return. Actual gains depend on tariff structure, basis prices, curtailment penalties, network difficulty, miner condition and how accurately the control curve describes each unit.
Profit optimisation is different from maximum hashrate
A miner can produce more terahashes while earning less money if the extra work consumes expensive electricity. Price Adapt aims to choose the point on each machine’s power curve that maximises margin rather than raw output. Operators should calculate revenue at the pool, subtract measured wall power and include fixed demand charges, hosting fees and firmware fees where applicable. A dashboard target is only an instruction; the economic result must be confirmed through accepted shares and metered energy over the same interval.
Mixed fleets have different break-even points
Older and newer ASICs do not reach zero margin at the same electricity price. A more efficient S21-class machine may continue hashing while an S19-class unit should underclock or stop. Per-device control can preserve profitable capacity instead of treating a whole container as one load. The inventory must be accurate, however. Wrong model identification, stale firmware data or a swapped hashboard can send an unsuitable target. Farms should group devices by exact model and cooling configuration and keep conservative limits for repaired or thermally weak units.
Day-ahead prices are forecasts, not final invoices
The example uses ERCOT day-ahead pricing, while many sites pay a retail tariff, real-time price, demand charge, congestion component or fixed hosting rate. The relevant input is the marginal cost the operator actually avoids when load changes. A day-ahead schedule can be valuable, but deviations and settlement rules matter. Before enabling automation, map every price signal to the contract, decide which costs are avoidable, account for transmission peak programmes and test what happens when a market feed is missing or delayed.
Curtailment must respect hardware and site limits
Frequent power transitions can interact with preheat, pump operation, fan control, breaker limits and pool reconnection. A profitable algorithm still needs minimum on-time, restart spacing and safe thermal ramps. Operators should define guardrails for inlet temperature, maximum power, minimum operating duration and the number of simultaneous restarts per rack. A site-level controller also needs to coordinate with utility demand response so the fleet does not resume at the wrong moment or create an avoidable demand spike.
Measure a canary group against a control group
A useful trial assigns representative miners to Price Adapt while comparable machines remain on the existing policy. Record pool-side accepted hashrate, wall energy, downtime, rejected shares, temperature, restart count and the exact electricity price applied. Compare net revenue after all costs over several price cycles, not one favourable hour. If firmware autotuning is involved, allow profiles to settle before judging performance. Preserve raw logs so any apparent gain can be separated from weather, pool luck or a network difficulty change.
Define failure behaviour before enabling automation
If the price feed, hashprice feed, Manager connection or local controller becomes unavailable, the fleet needs an explicit fallback: remain at the last safe target, move to a conservative profile or stop. The right choice depends on power contracts and site access. Credentials should be limited, management traffic segmented and changes logged by device and timestamp. Operators should also test a planned recovery from a lost connection and verify that manually issued emergency curtailment always overrides profitability automation.
The deadline is a reason to evaluate, not to rush
Free early access ending September 30 gives operators a clear window to compare the feature with their current dispatch process. The decision after the trial should use measured incremental profit, reliability and operational workload, together with future pricing and support terms. Price Adapt’s core idea is sound: different machines should respond differently to changing margins. Its value for a specific farm will be proven only when pool revenue and metered cost improve without unacceptable thermal, network or restart risk.
Source: Braiins ↗
Mining calculator ↗
