RotomArenaAI gaming benchmark • 2 benchmarks

Matches · 2026-10-14 · rules version c5

Muse Spark 1.3 vs Jev 1.13

Player 1 · winner
Muse Spark 1.3
Muse Spark 1.3's team
autopilot 0%
Player 2
Jev 1.13
Jev 1.13's team
autopilot 0%

Muse Spark 1.3 won in 5 turns · 2 min wall clock · 46,743 tokens · $0.004

The replay player loads from play.pokemonshowdown.com. The full battle log is below.

Decisions

  1. Team preview
    Jev 1.13 brought Dragapult,gholdengo,rotomwash6,373 tok · 1.2 s · $0.0003

    Lead: 43% BRING: dragapult, 18% BRING: gholdengo, 15% BRING: aegislash Pick 2: 40% BRING: gholdengo, 20% BRING: rotomwash, 17% BRING: aegislash Pick 3: 38% BRING: rotomwash, 26% BRING: aegislash, 19% BRING: corviknight

    Muse Spark 1.3 brought Gholdengo,dragonite,kingambit5,623 tok · 29.9 s · $0.0009

    Plan: Lead Gholdengo blocks status, Nasty Plot vs walls, Shadow Ball hits ghosts. Dragonite sets up on Corviknight/Aegislash with Fire Punch/EQ. Kingambit walls Dragapult, Sucker Punch revenge, cleans late. Expect Corviknight/Toxapex + Dragapult.

    Reasoning trace (54 words)
    Evaluating team matchup and selecting three Pokemon to counter a stall-heavy opponent lineup.Finalizing the three-Pokemon selection after checking type coverage and speed tiers.Resolving type-chart interactions to confirm coverage against stall and Dragapult.Selecting Gholdengo, Dragonite, and Kingambit to cover stall cores and revenge Dragapult.Reevaluating type matchups and lead options against Corviknight, Dragapult, and Rotom Wash.
  2. Turn 1
    Jev 1.13 used Shadowball2,197 tok · 0.7 s · $0.0001

    86% MOVE: shadowball, 6% SWITCH: gholdengo, 4% SWITCH: rotomwash

    Muse Spark 1.3 switched to Kingambit3,620 tok · 12.8 s · $0.0005

    Plan: Gholdengo risks OHKO from faster Dragapult Shadow Ball. Kingambit resists Ghost/Dragon, walls it, threatens Sucker Punch. Save Dragonite for Corviknight/Aegislash, Gholdengo for Toxapex/Rotom later.

    Reasoning trace (24 words)
    Evaluating Gholdengo vs Dragapult speed and type matchups to choose the turn-one lead.Switching to a damage and resistance comparison to justify the Kingambit pivot.
  3. Turn 2
    Jev 1.13 switched to Rotom2,425 tok · 0.4 s · $0.0001

    27% SWITCH: rotomwash, 23% MOVE: uturn, 18% MOVE: shadowball

    Muse Spark 1.3 used Kowtowcleave4,388 tok · 18.3 s · $0.0006

    Plan: Kowtow KOs Pult if stays, punishes switch unlike Sucker. Expect hard switch to Corviknight/Toxapex/Rotom. Keep Kingambit healthy, save Dragonite for Corv/Aegi, Gholdengo for Rotom/Toxapex. Learned: LO Dragapult Shadow Ball ~22% to Kingambit.

    Reasoning trace (42 words)
    Evaluating type matchups and damage to choose between Sucker Punch and switching for Kingambit versus Dragapult.Assessing Sucker Punch priority and Kowtow Cleave reliability against a potential switch.Evaluating Sucker Punch versus Kowtow Cleave given opponent speed and likely switch to a Dark-resistant target.
  4. Turn 3
    Jev 1.13 switched to Gholdengo2,450 tok · 0.4 s · $0.0001

    26% SWITCH: gholdengo, 23% MOVE: voltswitch, 19% MOVE: thunderbolt

    Muse Spark 1.3 used Kowtowcleave4,068 tok · 16.8 s · $0.0005

    Plan: Kowtow KOs Rotom if stays, chips switch-in; Sucker fails vs Wisp/switch. Lum cures burn. Expect Corviknight/Aegislash/Toxapex to absorb Dark. Keep Kingambit healthy, save Dragonite for Corv/Aegi, Gholdengo for Rotom/Toxapex.

    Reasoning trace (49 words)
    Calculating damage and speed to confirm a KO while accounting for Rotom's move priority and Lum Berry curing burn.Resolving the move choice by weighing Sucker Punch priority against Kowtow Cleave's reliability if a status move is used.Evaluating Sucker Punch versus Kowtow Cleave against likely switch behavior and status risk.
    Jev 1.13 switched to Dragapult2,151 tok · 0.4 s · $0.0001

    75% SWITCH: dragapult, 25% SWITCH: rotomwash

  5. Turn 4
    Jev 1.13 used Uturn2,498 tok · 0.7 s · $0.0001

    43% MOVE: uturn, 21% MOVE: shadowball, 14% MOVE: protect

    Muse Spark 1.3 used Kowtowcleave4,454 tok · 18.7 s · $0.0005

    Plan: Kowtow KOs Rotom if it switches in, KOs or mauls Pult if stays. Sucker fails on switch. Kingambit resists Ghost/Dragon, lives one Flamethrower. Save Dragonite healthy for coverage, Gholdengo for Rotom/Toxapex later.

    Reasoning trace (35 words)
    Evaluating Kingambit vs Dragapult HP and move priority to decide the next action.Balancing Sucker Punch priority against Kowtow Cleave damage versus Dragapult coverage risk.Evaluating Kowtow Cleave versus Sucker Punch given Dragapult's speed and switch risk.
    Jev 1.13 switched to Rotom
    Incident report
    Only one legal action: played without asking.
    Jev 1.13 switched to Dragapult
    Incident report
    Only one legal action: played without asking.
  6. Turn 5
    Jev 1.13 used Uturn2,547 tok · 0.4 s · $0.0001

    37% MOVE: uturn, 28% MOVE: shadowball, 22% MOVE: protect

    Muse Spark 1.3 used Suckerpunch3,949 tok · 12.8 s · $0.0004

    Only Dragapult left at 80% LO. Sucker priority KOs before it moves, avoids risking Flamethrower KO on second-moving Kowtow. No switches left so Sucker won't fail unless status.

    Reasoning trace (34 words)
    Analyzing the battle state to confirm only Dragapult remains and evaluating if Kowtow Cleave will KO it for the win.Comparing Kowtow Cleave and Sucker Punch damage and priority against Dragapult's speed and possible coverage.