Matches · 2026-10-10 · rules version c4
Gemini 3.8 Flash vs GLM 5.3 Flash
Gemini 3.8 Flash won in 7 turns · 4 min wall clock · 46,051 tokens · $0.025
The replay player loads from play.pokemonshowdown.com. The full battle log is below.
Decisions
- Team previewGLM 5.3 Flash brought
Gholdengo,sylveon,dragoniteOpponent has Mimikyu, Gholdengo, and Dragonite cores. Sylveon's Hyper Voice hits Mimikyu's disguise and Dragonite hard, Gholdengo blocks status and pressures Primarina, and Scarf Dragonite with Multiscale gives speed control and Ice Spinner for Garchomp. This trio covers their team best.
Reasoning trace (30 words)
Pick a solid core: Gholdengo, Sylveon, Dragonite? Opponent has Gholdengo blocking status; Sylveon Hyper Voice good vs Mimikyu/Dragonite. Bring Gholdengo, Sylveon, Kingambit? Kingambit weak to Primarina/Fighting... Bring Gholdengo, Sylveon, Dragonite.
Gemini 3.8 Flash broughtGholdengo,dragonite,mimikyuChoice Scarf Gholdengo provides fast offensive momentum and outspeeds opponent's Garchomp and Gholdengo. Dragonite offers great defensive synergy and setup potential with Multiscale, while Mimikyu acts as the ultimate safety net and revenge killer with Disguise.
- Turn 1GLM 5.3 Flash used
ShadowballShadow Ball is super effective against their Gholdengo and I likely outspeed or at least trade well. Make It Rain is resisted and self-drops SpA, while Nasty Plot wastes a turn. Attacking now pressures their lead immediately.
Reasoning trace (9 words)
Shadow Ball is 2x on Gholdengo, likely big damage.
Gemini 3.8 Flash usedShadowballChoice Scarf guarantees our Gholdengo outspeeds opposing Gholdengo unless it also runs Scarf and wins a speed tie. Shadow Ball is super-effective STAB that will either knock it out or leave it critically weak.
GLM 5.3 Flash switched toDragoniteTheir Gholdengo is faster than mine was and just KO'd my lead. Scarf Dragonite outspeeds it and Earthquake hits its Steel typing hard, likely KOing before it can act.
Reasoning trace (10 words)
Gholdengo mirror; Dragonite scarf outspeeds and EQ hits Steel. Dragonite.
- Turn 2GLM 5.3 Flash used
EarthquakeEarthquake is super effective on Gholdengo and Dragonite outspeeds most variants with Choice Scarf, likely scoring a knockout. Extremespeed is immune-blocked and Dragon moves are resisted.
Gemini 3.8 Flash usedShadowballGholdengo is Choice Scarf-locked into Shadow Ball and outspeeds Dragonite. Attacking safely breaks Multiscale and deals solid chip damage before Dragonite can retaliate.
- Turn 3GLM 5.3 Flash used
EarthquakeEarthquake already left Gholdengo at 22%; another Earthquake KOs it before it can do anything meaningful. Dragonite is faster with Choice Scarf in most cases, and even if outsped, Multiscale is broken but 22% Gholdengo's Shadow Ball won't KO from 75%.
Gemini 3.8 Flash usedShadowballGholdengo is locked into Shadow Ball and outspeeds Dragonite. With Dragonite's Multiscale broken and Special Defense at -1, another Shadow Ball will knock it out or leave it critical, while Gholdengo is immune to Extreme Speed.
Gemini 3.8 Flash switched toDragonite - Turn 4Gemini 3.8 Flash used
ExtremespeedOpposing Dragonite is at 6% HP. Extreme Speed has +2 priority, ensuring we move first and safely KO it without risking damage to our Multiscale.
GLM 5.3 Flash switched toSylveonEarthquake is immune against Dragonite, and Dragon Claw into full-HP Multiscale won't KO while our 6% Dragonite dies to any hit. Switching to Sylveon is the only safe option.
Reasoning trace (22 words)
Dragonite mirror, we're at 6% with -2 SpD, Multiscale broken. Their Dragonite full HP with Multiscale. Earthquake does nothing. Switch to Sylveon.
- Turn 5Gemini 3.8 Flash used
EarthquakeGLM 5.3 Flash usedHypervoiceHyper Voice is Pixilate-boosted Fairy STAB, hitting Dragonite for 2x damage — roughly half through Multiscale, potentially a KO without it. Sylveon resists Extreme Speed and tanks it easily, so there's no need to pivot; chipping Multiscale off now sets up a clean KO next turn.
Reasoning trace (1,209 words)
Let me analyze this situation. My active: Sylveon at 138/202 (68.3%), leftovers, pixilate. Moves: Hyper Voice (pixilate makes it Fairy, 90 BP × 1.2 = 108 effective), Wish, Protect, Yawn. Opponent's active: Dragonite at 100% HP, unknown item, ability unknown (innerfocus or multiscale). Revealed move: Extreme Speed. It used Extreme Speed last turn on Sylveon, dealing 126→138... wait, Sylveon came in at some HP, took Extreme Speed, went to 126, then healed to 138 with leftovers. So Extreme Speed did some damage. Speed: Sylveon 80 vs Dragonite 90-145. Dragonite likely moves first. Dragonite has Extreme Speed priority anyway. Key considerations: - Sylveon is Fairy, so Dragonite's Extreme Speed (Normal) is not very effective (0.5x). Dragonite's other moves likely: Dragon Claw/Outrage (Dragon, super effective vs Fairy? No wait, Dragon doesn't affect Fairy at all — 0x). Dragonite's STABs: Dragon (immune vs Fairy) and Normal/Flying. Flying moves like Hurricane/Aerial Ace would be neutral vs Fairy. Earthquake neutral. - Sylveon's Hyper Voice (Fairy via pixilate) is super effective vs Dragonite (Dragon weak to Fairy, 2x). Dragon/Flying: Fairy 2x, Ice 4x, Dragon 2x, Rock 2x. So Hyper Voice hits Dragonite super effectively. Sylveon SpA 178, Hyper Voice effective BP 108 with pixilate. Dragonite at full HP — if multiscale, damage halved. Let's estimate: 178 SpA, 108 BP, vs Dragonite SpD ~108-167, HP 166-198. Damage calc rough: Level 50, ((2*50/5+2)=22 * BP 108 * SpA 178 / SpD ~140)/50 + 2 = 22*108*178/140/50 + 2 = (22*108=2376; 2376*178=422,928; /140=3020.9; /50=60.4; +2=62.4) then ×1.5 STAB... wait pixilate already gives 1.2x power but Sylveon still gets STAB on Fairy move: 1.5x. So 62.4 × 1.5 = 93.6, ×2 (super effective) = 187. That's more than Dragonite's max HP (198). Wait, that seems too high. Let me recompute. Damage formula: (((2*Level/5 + 2) * Base * A/D)/50 + 2) * modifiers 2*50/5+2 = 22. Base = 90 × 1.2 (pixilate) = 108. A = 178, D = let's say 140 (mid). 22 * 108 = 2376. 2376 * 178 = 422,928. 422,928 / 140 = 3020.9. /50 = 60.4. +2 = 62.4. STAB 1.5 → 93.6. SE 2x → 187.2. Random 0.85-1.0 → 159-187. Dragonite HP max 198. So even with multiscale halving (if full HP), 79-93 damage out of ~166-198 HP. That's ~40-56%. Not a KO through multiscale, but a big chunk. Without multiscale (if inner focus), it could KO or nearly KO: 159-187 vs 166-198 HP — could be a KO on high rolls vs min HP, but not guaranteed. But wait — Dragonite moves first (or with Extreme Speed priority). Dragonite's Extreme Speed vs Sylveon: Normal vs Fairy = 0.5x. Last turn it did 202-126... wait Sylveon came in — we don't know entry HP. Sylveon was at full 202 presumably when it came in, took E-speed to 126, so 76 damage? That seems like a lot for resisted. Hmm, wait: "Your Sylveon came in. Opposing Dragonite used Extreme Speed. Your Sylveon is at 126/202 HP." So E-speed did 76 damage through resistance? That would require a very strong Dragonite with a band or something. Actually 76 damage resisted: base 80, A ~204 max, D 152. 22*80*204/152/50+2 = 22*80=1760; *204=359,040; /152=2362; /50=47.2; +2=49.2; STAB 1.5=73.8; resisted 0.5=36.9. That's only ~37. To do 76, it'd need... hmm, maybe Sylveon didn't come in at full HP? Or maybe crit? No crit mentioned. Or maybe Sylveon took entry damage? No hazards. Hmm, maybe Sylveon came in at less than full? Actually it says "came in" — switching in, should be full HP unless it was damaged earlier. Sylveon wasn't mentioned before turn 4, so it was on bench... wait, actually the bench shows only gholdengo (fainted) and dragonite. My team: Sylveon active, Gholdengo fainted, Dragonite 6% on bench. So I have 3 Pokémon: Sylveon, Gholdengo (dead), Dragonite (6%). Hmm wait, actually maybe Sylveon took 76 from E-speed because Dragonite has a band and max attack: 22*80*204/152... let me redo with D=152: 22*80 = 1760, *204 = 359,040, /152 = 2362.1, /50 = 47.24, +2 = 49.24, ×1.5 = 73.9, ×0.5 = 36.9. Still ~37. So 76 damage doesn't add up unless Sylveon entered with less HP or there's something else. Maybe Sylveon was chipped earlier? The log only shows turns 2-4. Turn 1 unknown. Maybe Sylveon was active turn 1 and took damage. Anyway. Actually wait — maybe the opponent's Dragonite is Choice Band with max Atk and rolled high: 36.9 × 1.0... no. Or maybe it's not resisted... Extreme Speed is Normal, Sylveon is Fairy, resisted 0.5. Hmm. Or maybe Sylveon came in at 202, took E-speed (37), then... 202-37 = 165, not 126. So 76 damage. Something's off. Maybe Sylveon took prior damage in turn 1-2. Whatever — the point is Sylveon is at 138/202. Threat assessment: Dragonite at full HP with unknown ability (multiscale likely since it's listed). If multiscale, Hyper Voice does ~40-55%. Dragonite then hits back with E-speed (resisted, ~37) or something else. Sylveon can tank E-speed fine. Options: 1. Hyper Voice: chunk Dragonite hard (through multiscale ~40-55%, without ~80-95%+). Sylveon tanks E-speed (resisted ~37). Next turn another Hyper Voice likely KOs (multiscale broken). 2. Wish: heal 101 next turn, but Dragonite gets a free turn — could set up Dragon Dance? It's choice-scarfed? No wait, my Dragonite has scarf. Opponent's Dragonite item unknown. If opponent's Dragonite is Banded, it's locked into E-speed? It used E-speed turn 4. If choice-locked into E-speed, that's great for us — E-speed is resisted. But unknown. 3. Protect: scout, but passive. 4. Yawn: put Dragonite to sleep in 2 turns, but Dragonite could just KO Sylveon? No — Sylveon tanks E-speed. Yawn then Dragonite must switch or fall asleep. But Yawn is Normal status — Good as Gold is gone (opponent's Gholdengo fainted). Yawn works. But Dragonite might switch out, wasting it. Also inner focus doesn't block sleep. Yawn is slow though — Sylveon is slower, Dragonite acts first. 5. Switch to Dragonite (6% HP): terrible, it dies to anything. Best play: Hyper Voice. It's super effective, chunks Dragonite hard, breaks multiscale potentially (if it does damage, multiscale breaks since Dragonite won't be at full HP). Sylveon is bulky and resists E-speed. Even if Dragonite has Dragon Dance or something, Sylveon's fairy STAB threatens it. Wait — could Dragonite OHKO Sylveon with something? Dragon moves immune (Fairy). Flying moves neutral: Hurricane 110 BP special... Dragonite SpA lower. E-speed resisted ~37. Sylveon at 138 + leftovers. Sylveon is fine. Could Dragonite be faster and... it doesn't matter much; Sylveon tanks whatever. Actually, one consideration: if opponent's Dragonite has Multiscale and full HP, Hyper Voice does ~40-55%. Then Dragonite E-speeds (~37 resisted), Sylveon at ~101+6 leftovers. Next turn Hyper Voice KOs (multiscale broken, need ~50% more). That's a clean 2-turn KO path while Sylveon stays healthy. Alternatively Yawn: Dragonite falls asleep in 2 turns unless it switches. But Hyper Voice progress is more concrete, and Sylveon's Hyper Voice threatens immediate heavy damage. Also if Dragonite is banded/locked into E-speed, we win the matchup entirely. Hyper Voice is the clear best action. One more check: could Hyper Voice KO outright if no multiscale? 159-187 vs 166-198 HP — high rolls KO low-HP Dragonite. Even if not, big chunk. And if multiscale, still ~45%. Either way best. Decision: MOVE: hypervoice
- Turn 6Gemini 3.8 Flash used
ExtremespeedExtreme Speed did 38% to Sylveon on turn 4, which easily finishes its remaining 29% HP with priority, guaranteeing the KO without risking a Quick Attack or speed tie.
GLM 5.3 Flash usedHypervoiceautopilotReasoning trace (2,221 words)
Let me think about this turn. My Sylveon is at 60/202 (29.7%). Opponent's Dragonite is at 45% HP, faster than Sylveon (90-145 vs 80). Dragonite has Extreme Speed (priority) and Earthquake. If I stay in, Dragonite will likely KO Sylveon with Extreme Speed (priority move, Sylveon at 29.7%). Sylveon is Fairy, so Extreme Speed is neutral (Normal vs Fairy = 0.5x actually — Normal is not very effective against Fairy? No wait, Normal is resisted by... Rock and Steel. Fairy is neutral to Normal. Actually Normal-type moves are not very effective against Rock and Steel. Fairy takes neutral from Normal.) So Extreme Speed at 29.7% HP Sylveon — Dragonite's attack 138-204, ESpeed BP 80, likely KOs Sylveon from 29.7%? Sylveon's Def is 85, which is low. Dragonite Atk ~170 avg, ESpeed 80 BP neutral — that's likely a KO from 29.7%. Yes, probably KOs. Options: 1. Hyper Voice: Sylveon is slower, so Dragonite moves first. If Dragonite KOs Sylveon with ESpeed, Hyper Voice never happens. If Sylveon survives, Hyper Voice does ~45% → could KO Dragonite (it's at 45%). Hyper Voice previously did damage taking Dragonite from full to 45% in one hit (super effective — wait, Fairy vs Dragon is super effective, yes 2x). So another Hyper Voice would KO Dragonite if Sylveon survives. 2. Wish: Sylveon likely dies to ESpeed first. Bad. 3. Protect: Sylveon survives this turn (unless Dragonite uses something else... Protect blocks everything). Then next turn Sylveon at 60 + leftovers ~72, still in KO range of ESpeed. But Wish could be used next turn... Actually Protect then Wish then... Sylveon is slow and frail on physical side. 4. Yawn: Sylveon likely dies before Yawn matters (Yawn puts to sleep after 1 turn, but Sylveon must survive). 5. Switch to Dragonite (6% HP): Dragonite at 6% with Multiscale broken (not full HP). Opponent's Dragonite could ESpeed it? Switching means the incoming Dragonite takes the attack. Opponent's Dragonite ESpeed vs my 6% Dragonite — KOs. Bad. So the real choice: Hyper Voice (gamble on surviving ESpeed) vs Protect. Let me estimate: Dragonite Atk max 204, min 138. Level 50, ESpeed 80 BP, neutral vs Fairy. Sylveon Def 85, HP 202, current 60. Damage = ((2*50/5+2)*80*Atk/Def)/50+2 = (22*80*Atk/85)/50+2 = (1760*Atk/85)/50+2 = 20.7*Atk/85... let me compute: 22*80 = 1760. 1760 * Atk / 85. If Atk = 170: 1760*170 = 299,200 / 85 = 3520. /50 = 70.4 + 2 = 72.4. Then *0.85-1.0 variance → 61-72 damage. Sylveon at 60 HP. So even minimum roll (61) KOs. Even Atk 138: 1760*138/85 = 2857/50 = 57 + 2 = 59. *0.85 = 50. Hmm, min Atk 138 gives 50-59 damage, Sylveon at 60 could survive minimum roll. But Dragonite's Atk range 138-204 — a typical Dragonite is Adamant with investment, so Atk likely ~180-204. That KOs Sylveon with ESpeed almost certainly. So Hyper Voice = Sylveon dies, Dragonite survives at 45%. Then I bring in Dragonite (6%) vs their Dragonite (45%) — my Dragonite dies to ESpeed probably (6% HP, ESpeed priority). Actually my Dragonite at 6% — any hit KOs. Hmm, that's bad. Let me think about the whole game state. My team: Gholdengo fainted, Dragonite 6%, Sylveon 29.7%. Opponent: Dragonite 45%, Gholdengo fainted, one more to come from Mimikyu, Garchomp, Volcarona, Primarina. Wait — "The opponent brought 3 of the 6 it showed at team preview: 1 more is still to come in" — so opponent has 2 revealed (Gholdengo fainted, Dragonite active) and 1 unrevealed. I have 3: Gholdengo fainted, Dragonite 6%, Sylveon active. So it's essentially 2v2 now: my Sylveon (29.7%) + Dragonite (6%) vs their Dragonite (45%) + unknown (full HP). If Sylveon dies this turn to ESpeed while using Hyper Voice... wait, if Sylveon uses Hyper Voice and dies to ESpeed first, no damage dealt. Then my 6% Dragonite comes in vs their 45% Dragonite. My Dragonite is Choice Scarf, speed 145. Their Dragonite speed 90-145. My Dragonite has ESpeed too (8 PP). My Dragonite at 6% — ESpeed from their Dragonite KOs it. But if I switch in and my Dragonite is faster (scarf 145 vs their possibly ≤145), my Dragonite could ESpeed first and KO their 45% Dragonite? My Dragonite Atk 186, ESpeed 80 BP vs Dragonite (Flying, neutral). Damage: 22*80*186/Def(~130)/50+2 = (1760*186/130)/50+2 = (327,360/130)/50+2 = 2518/50 = 50.4+2 = 52.4, *0.85-1 = 44-52. Their Dragonite HP 166-198, at 45% = ~75-89 HP. So ESpeed does ~44-52, not a KO. Then their Dragonite ESpeeds back and KOs my 6% Dragonite. Bad. Alternatively, Sylveon Protect: survives turn. Then next turn, what? Sylveon at 60+6=66ish. Dragonite ESpeeds again — 50-72 damage, likely KOs (66 HP, most rolls KO). Wish first? Wish then they ESpeed — Sylveon at 60, takes 50-72, likely dies before Wish activates. Hmm. Actually wait — can Sylveon survive? 60 HP. ESpeed damage range vs Sylveon: with Atk 138 (min): 50-59. With Atk 204 (max): 22*80*204/85 = 359,424/85 = 4228/50 = 84.6+2 = 86.6 → 73-86. So damage 50-86 depending on Atk. Sylveon at 60 survives only vs low-Atk Dragonite with low rolls. Most likely KO. Hmm, but what about Yawn? Sylveon is slower, dies to ESpeed before Yawn matters. Yawn needs Sylveon to survive the turn, then Dragonite gets drowsy, falls asleep at end of next turn (unless switches). If Sylveon survives with Yawn... but Sylveon likely dies. What's the best line? Let me think about Hyper Voice more carefully. If Sylveon survives ESpeed (unlikely but possible vs min Atk rolls), Hyper Voice KOs Dragonite (45% HP, Hyper Voice did ~55% last time from full... wait it took Dragonite from 100% to 45%, so ~55% damage. Another Hyper Voice = KO). Actually, let me reconsider: last turn Hyper Voice took Dragonite from full to 45%. So Hyper Voice does ~55% of their Dragonite's HP. Dragonite at 45% → Hyper Voice KOs. So the gamble: Sylveon Hyper Voice. If Sylveon survives ESpeed (low probability), Dragonite is KO'd, and I'm in great shape (Sylveon vs unknown). If Sylveon dies, I have 6% Dragonite vs 45% Dragonite + unknown — nearly lost. Alternative: Protect. Sylveon survives for sure. Next turn: Dragonite at 45%, Sylveon at ~66. If Sylveon uses Wish next turn, Dragonite ESpeeds — 50-86 damage vs 66 HP — likely KO, Wish wasted (well, Wish activates end of turn but Sylveon fainted... actually if Sylveon faints, Wish is gone). If Sylveon Hyper Voices next turn at 66 HP — dies to ESpeed first (slower), no damage. Hmm, same problem. Actually wait — what if Dragonite doesn't always click ESpeed? It might click Earthquake (neutral vs Fairy, 100 BP, no STAB... Dragonite's EQ vs Sylveon: 22*100*Atk/85/50+2, Atk 170: 22*100*170 = 374,000/85 = 4400/50 = 88+2 = 90 → 76-90 damage. Also KOs). Either way Sylveon likely dies. Hmm, so realistically Sylveon dies this turn or next unless protected. Protect delays one turn but then Sylveon still dies next turn (66 HP, ESpeed 50-86). Unless... Wish + Protect alternation? Wish (turn 6... no wait). Let me re-think: Turn 6 (now): Protect. Sylveon → 66 HP end of turn. Turn 7: Wish. Dragonite attacks (ESpeed 50-86 vs 66 HP — likely KO). Sylveon dies, Wish fails. Bad. Turn 7: Hyper Voice — Sylveon dies first (slower), no damage dealt. Bad. So Sylveon is basically dead next turn regardless. The question is whether to take Dragonite down with it. Hyper Voice now: ~15-25% chance Sylveon survives ESpeed (needs low Atk Dragonite + low roll), and if survives, KOs Dragonite. Expected value: small chance of great outcome. Actually, let me reconsider the damage calc. Sylveon's SpA 178, Hyper Voice 90 BP * 1.2 (Pixilate) = 108 effective, STAB 1.5 → 162 effective BP. Vs Dragonite SpD 108-167, 2x super effective (Fairy vs Dragon). Damage = (22 * 108 * 178 / SpD)/50 + 2, *1.5 STAB... wait I need to be careful. Formula: Damage = (((2*Level/5+2) * Power * A/D)/50 + 2) * modifiers. Level 50: 2*50/5+2 = 22. Power = 90*1.2 = 108 (Pixilate). A = 178. D = SpD. Base = 22*108*178/D/50 + 2 = (422,688/D)/50 + 2 = 8453.76/D + 2. If SpD = 130 (neutral nature no investment... Dragonite base SpD 100... wait Dragonite base SpD is 100? Dragonite: 91/134/95/100/100/80. At level 50, min SpD = 100... base 100 → min stat at level 50 = (2*100+0+0)*50/100+5 = 105? Formula: ((2*Base + IV + EV/4) * Level/100) + 5 = (2*100)*0.5+5 = 105 min, max = (2*100+31+63)*0.5+5 = 165. Hmm the given range says SpD 108-167, close enough. If SpD = 135: 8453.76/135 = 62.6 + 2 = 64.6. *1.5 STAB = 96.9. *2 SE = 193.8. *0.85-1.0 = 165-194. Dragonite HP 166-198, at 45% = 75-89. Way more than enough — KO for sure. OK so Hyper Voice KOs Dragonite if it goes through. Now the key question: does Sylveon survive ESpeed? Sylveon 60 HP. ESpeed damage 50-86 depending on opponent's Atk and rolls. If opponent's Dragonite is Adamant max Atk (204): 73-86 — guaranteed KO. If Jolly: Atk ~185: 22*80*185/85 = 325,600/85 = 3830/50 = 76.6+2 = 78.6 → 67-78 — guaranteed KO. If no Atk investment (Atk 138-150): 50-62 — Sylveon survives some rolls. A Dragonite running ESpeed + EQ is physical, almost certainly Adamant/Jolly with investment. So Sylveon dies to ESpeed ~always. Hmm wait, but actually — does the opponent know Sylveon is at 29.7%? They see HP bar. They'll click ESpeed for sure (it KOs and it's priority, avoiding speed issues). So Hyper Voice = Sylveon faints without dealing damage (95%+ of the time). That leaves 6% Dragonite vs 45% Dragonite + unknown full-HP mon. Nearly unwinnable. Protect = Sylveon survives this turn guaranteed. Then next turn Sylveon likely dies anyway. Same end state, one turn later. But wait — is there any line where Sylveon survives two turns? 66 HP after Protect + Leftovers. ESpeed min damage 50 (vs min Atk). Vs realistic Atk 185+: 67-78 — KOs 66 HP. So no. Hmm, what about the possibility the opponent switches? Their Dragonite at 45% might switch out to the unknown mon to preserve it? If I Protect and they switch, I get a free turn... but Sylveon still weak. Actually, let me think about it differently. What's my win condition? My Dragonite at 6% with Choice Scarf, ESpeed, Dragon Claw, EQ, Icespinner. It's nearly dead — dies to any hit. Its ESpeed does 44-52 to their Dragonite (calculated above) — not a KO from 45% (~75-89 HP). So my Dragonite can't KO their Dragonite in one hit. Their Dragonite ESpeed KOs my 6% Dragonite trivially. So realistically: their Dragonite beats both my remaining mons unless Sylveon can KO it. Sylveon's Hyper Voice KOs it. The only way Hyper Voice connects is if Sylveon survives ESpeed — very unlikely. Alternatively — Yawn! If Sylveon survives (unlikely), Yawn puts Dragonite to sleep next turn. But Sylveon must survive, same issue. Hmm, wait. Actually, let me reconsider whether Sylveon survives. Let me redo the ESpeed calc more carefully. Sylveon: HP 202, Def 85. Current HP 60. Dragonite Atk range 138-204 (given). ESpeed BP 80, neutral vs Fairy (Normal vs Fairy = 1x, yes neutral). Damage = ((22 * 80 * Atk / 85) / 50 + 2) * random(0.85-1.0) Atk=138: 22*80 = 1760; 1760*138 = 242,880; /85 = 2857.4; /50 = 57.1; +2 = 59.1; range 50-59. Sylveon at 60 survives all rolls! Atk=150: 1760*150 = 264,000; /85 = 3105.9; /50 = 62.1; +2 = 64.1; range 54-64. Survives rolls ≤59 (~42% of rolls... 54-59 out of 54-64, 6/11 ≈ 55% survive). Atk=170: 72.4 → 61-72. Survives only if roll ≤ 60/72... damage must be ≤59 to survive (60 HP survives damage ≤59). 61-72 range: never ≤59. Dies always. Atk=204: 86.6 → 73-86. Dies always. So Sylveon survives only if opponent's Dragonite has uninvested Attack (Atk ≤ ~150) AND low rolls. A Dragonite with ESpeed/EQ is physical — Atk investment near certain. So Sylveon dies. OK here's another thought: what about the unknown incoming Pokémon? The opponent still has 1 unrevealed from Mimikyu, Garchomp, Volcarona, Primarina. If their Dragonite is at 45% and they predict my Hyper Voice... they can't really pivot meaningfully. Let me also consider: is there value in Protect to scout? Not really — we know their moves. What about switching Sylveon out to Dragonite? Then Sylveon keeps 60 HP, Dragonite (6%) comes in and eats ESpeed — dies. Terrible. So the options boil down to: A) Hyper Voice: likely Sylveon dies without effect; small chance (opponent's Dragonite weak Atk) Sylveon survives AND KOs Dragonite. B) Protect: Sylveon survives now, dies next turn to ESpeed (66 HP vs 67-78+ damage). No progress. Actually wait — could Sylveon survive ESpeed at 66 HP if opponent's Atk is low? Same analysis shifted: needs damage ≤65. Atk 138: 50-59 — survives. Atk 150: 54-64 — survives most. Atk 170: 61-72 — survives rolls 61-65 (~38%). Hmm interesting, at 66 HP there's a slightly better survival chance vs mid Atk. But even if Sylveon survives turn 7, what does it do turn 7? If it Hyper Voices turn 7 (after Protect turn 6), it dies to ESpeed before moving (slower) — unless it survives. Ugh, this is circular: Sylveon is slower and will die to priority ESpeed before acting, every turn, unless it Protects. So the sequence: Protect (T
Incident report
[PARSE ERROR] answer cut off at the 4096-token budget (missing <decision>...</decision> block). Default move played. Raw answer:
GLM 5.3 Flash switched toDragoniteSylveon fainted and my only remaining Pokémon is Dragonite at 6% HP, so I'm forced to send it in. Any hit from the opposing Dragonite will likely finish it, but there's no other option.
Reasoning trace (150 words)
Let me analyze this situation. My active Pokémon: Sylveon just fainted. My bench: - Gholdengo: fainted - Dragonite: 10/166 HP (6%), Choice Scarf Opponent's active: Dragonite at 45% HP, with Extreme Speed and Earthquake revealed. The only legal action is SWITCH: dragonite (6.0% HP). So I must switch in my Dragonite at 6% HP. That's the only legal action anyway. Wait, the format is bring 6 pick 3. I brought... let me think. My bench shows Gholdengo fainted, Dragonite at 6%. Sylveon fainted. So I have only Dragonite left? That means I brought 3: Sylveon, Gholdengo, Dragonite. All that's left is Dragonite at 6% HP. So I have to switch in Dragonite. It's the only legal action. Thought: Sylveon fainted and Dragonite is my last Pokémon, so I must send it in. It's at 6% HP, so any hit will KO it, but there's no other choice. Decision: SWITCH: dragonite
- Turn 7GLM 5.3 Flash used
ExtremespeedI'm at 6% HP, so any hit kills me. Opposing Dragonite at 45% will likely Extreme Speed; my Choice Scarf makes me faster, so my priority ESpeed lands first and KOs before theirs. Dragon Claw risks being outsped and losing.
Reasoning trace (46 words)
Dragonite at 6% HP, opposing Dragonite 45%. Extreme Speed priority wins the race—opponent likely uses ESpeed too; speed tie? Scarf makes us faster but ESpeed priority both. If both ESpeed, faster one first; ours likely faster with scarf. Dragon Claw could be outsped. ESpeed is safest.
Gemini 3.8 Flash usedExtremespeedOpposing Dragonite is down to 6% HP and Extreme Speed has +2 priority, guaranteeing a KO before it can move.