The reported one-fight loadout

The reference describes four fully condensed Omascul, a target Pal fed Seafood Salad, the Growth Acceleration Bell, and one normal Zenara/Astralym tower clear. It also reports a two-Omascul/three-trainee route over three fights. These are community test reports, not runs reproduced by Palworld Run.

Why we do not label the result verified

The 1.0.3.101283 native EXP consumer supports separate eligible Omascul components adding their bonuses even though every raw rank row retains a false same-species stacking flag. This resolves the apparent flag-versus-party-loop conflict for this route. It does not verify the source’s complete one-fight result: exact boss/event rewards, level normalization, effective settings and the starting EXP still matter.

A controlled way to test the exception

  • Record build, world EXP setting, starting player/Pal levels and exact target EXP.
  • Use the same boss and reset process with one Omascul, then two, with identical ranks and no food/accessory changes.
  • Compare actual EXP awarded before testing four copies. Add the bell and food one at a time.
  • Count trainee slots and complete reset time; a faster single trainee may still lose to a larger batch.

A reliable planning fallback

Choose the number of Omascul and trainees that keeps the clear safe and repeatable. Extra eligible Omascul components can raise the shared Pal factor while taking slots away from trainees. Compare complete batch times and actual EXP before choosing a one-trainee or larger-batch setup; a one-fight outcome remains unverified.

How party EXP bonuses combine

For party Pals, start the shared factor at 1 and add each eligible component’s EXP rate minus 1. Include the player’s component with equipped-item effects enabled. Multiply that shared factor by the recipient Pal’s own active EXP-food rate, the world EXP rate and the event or recipe rate before level normalization. Feeding the player does not replace feeding the trainee.

For example, if two eligible rank-5 Omascul components each return 1.8 and the equipped Bell makes the player component return 1.5, the shared factor is 1 + 0.8 + 0.8 + 0.5 = 3.1. A trainee with Seafood Salad’s 1.2 food rate then has a 3.72 pre-normalization rate at world and event rates of 1. This conditional calculation is not final awarded EXP or a level prediction.

The player receives a separate award before the shared Pal bonus is collected. That player branch uses the player’s own food and any applicable Relic EXP bonus; Omascul and the Bell’s later shared Pal factor do not multiply that preceding player award.

The Bell is equipped in the armor/accessory inventory read by this route, not activated by keeping a Key Item. This reader uses armor inventory type 4 and includes each distinct item FName once. Two equipped copies with the same Bell item ID do not supply two Bell contributions. This establishes neither arbitrary accessory-tier combinations nor all other equipment interactions.

Why a multiplier does not predict a level

Party Pal EXP still passes through level normalization. Its native setting registers a default of enabled, but a live setting override has not been observed. The enabled path uses recipient/player levels, level-difference settings and the current EXP table in level segments, with rounding inside that process. The final award sink has its own integer conversion and minimum award; these stages cannot be replaced by one universal rounding rule.

The configured EXP database points to the current Pal EXP table. Its separate migration-V1 reference is not used in the inspected normalization path. Exact event reward inputs, starting EXP, effective world settings and remaining inherited level settings are required for a numerical prediction. A one-fight 1→80 result, a fastest loop and creator timing still require their own reproducible run evidence.

Native food units, replacement and active time

Food uses three independent channels: ordinary meal, HP regeneration and keep-full. The latest food replaces its own channel, even if weaker; eating the same food refreshes that channel to its full recorded duration. Ordinary meals do not stack with other ordinary meals. The three channels can coexist.

Durations count eligible natural updates. The acquired default player, mounted Pal, deployed AI-controlled Pal and basecamp paths update once per active game-time second. Timer phase, catch-up and native eligibility guards apply; these are not universal wall-clock expiry forecasts. Timing for stored or unsummoned individuals without a valid pawn is not established.

Attack, Defense and Work speed use separate food stat factors; ordinary passive sums are a different stage. EXP food follows the recipient in the traced party-award paths. Direct base-worker crafting awards use a separate route without this food lookup.

Recovery Meds restore flat 3/5/10/15 HP per eligible one-second activation, capped at effective maximum HP and requiring positive current HP. They do not revive. Nutrient Tonic stops the traced satiety drain through its own channel. Salt-Grilled Skutlass enables leanback/knockback resistance checks; its raw 1 is not a percentage chance. Seafood Pasta has zero eligible normal explosion damage factor, with the original damage guards.

The no-effect JamBun row has duration 0 and returns before overwriting a channel. It does not clear an existing meal. Food application does not establish an item recipe or availability.