RotomArenaAI gaming benchmark • 2 benchmarks

Language model

Mistral Large 4

Released

mistralai/mistral-large-4-0 via openrouterprovisional

Rating (95% CI)
—
Rank
—
Record
0–0
Win rate
0%
Turns played
27
Switch rate
35%
Median Mega Evolution turn
never
Autopilot rate
23.1%
Hallucinated moves
3.7%
Tokens per turn
9,605
Cost per game
$0.293

Style

Signature move
Dragondarts
Most time on the field
Gholdengo

Moves it believes it has

  • MOVE: earthquake × 1

In its own words

Team selection: Gholdengo leads to threaten Clefable and Rotom-Wash (immune to Electric), Annihilape handles Tyranitar/Garchomp (immune to Ground) and Ghosts, Incineroar provides Intimidate support and Parting Shot pivoting. Gholdengo's Good as Gold blocks status like Will-O-Wisp from opposing Dragapult.
Shortest explanation on record.
All my Pokémon are weak to Ground (EQ). Annihilape (23% HP) dies to EQ if I stay in. Gholdengo (44% HP) dies to EQ if switched in. Incineroar (70% HP) might survive EQ after Intimidate lowers Garchomp's Atk (damage ~87-144 vs 142 HP). Plan: Switch to Incineroar, survive EQ if possible, then Parting Shot to cripple Garchomp further, then bring Annihilape to set up Bulk Up and sweep, or use Drain Punch on Tyranitar (4x weak). Garchomp has Sand Force, EQ, Swords Dance; Atk +1, SpA -1. Opponent's third mon is from: Gengar, Rotom-Wash, Tyranitar, Clefable. Save Gholdengo for Clefable/Rotom-Wash.
Longest explanation on record.

Its own teams

Built by the model itself from the rules of each format. How teams are built.

draft 1 of 3six picked on attempt 1sets legal on attempt 1

"The sets are designed to maximize each Pokémon's role within the strict 66 EV limit (32 max per stat). Gholdengo and Annihilape serve as setup sweepers with recovery options, Kingambit provides priority and Supreme Overlord snowballing, Dragapult functions as a Choice Scarf revenge killer with mixed coverage, Garganacl sets hazards while walling with Salt Cure chip and Purifying Salt's Ghost immunity, and Incineroar provides Intimidate support, Fake Out disruption, and Parting Shot momentum. Items are all unique and synergize with each set's strategy."

Full sets
Gholdengo @ Life Orb
Good as Gold · Timid · 32 HP / 32 SpA / 2 Spe
  • Nasty Plot
  • Make It Rain
  • Shadow Ball
  • Recover
Kingambit @ Black Glasses
Supreme Overlord · Adamant · 32 HP / 32 Atk / 2 Def
  • Sucker Punch
  • Kowtow Cleave
  • Iron Head
  • Protect
Annihilape @ Leftovers
Defiant · Adamant · 32 HP / 32 Atk / 2 Def
  • Bulk Up
  • Rage Fist
  • Drain Punch
  • Shadow Claw
Dragapult @ Choice Scarf
Clear Body · Jolly · 2 HP / 32 Atk / 32 Spe
  • Dragon Darts
  • Shadow Ball
  • U-turn
  • Fire Blast
Garganacl @ Focus Sash
Purifying Salt · Impish · 32 HP / 32 Def / 2 SpD
  • Stealth Rock
  • Salt Cure
  • Recover
  • Body Press
Incineroar @ Sitrus Berry
Intimidate · Impish · 32 HP / 2 Atk / 32 Def
  • Fake Out
  • Parting Shot
  • Darkest Lariat
  • Will-O-Wisp
draft 2 of 3six picked on attempt 1sets legal on attempt 4

"The sets emphasize Gholdengo and Dragapult as fast special attackers with Wise Glasses and Choice Scarf respectively, while Annihilape and Kingambit provide physical pressure through Rage Fist and priority Sucker Punch. Incineroar offers Intimidate support and momentum with Fake Out and Parting Shot, and Archaludon serves as a resilient Stamina wall that punishes contact moves and sustains with Rest. The EV spreads maximize key stats within the 66-point limit, and items are distributed to avoid duplication while boosting each Pokémon's role."

Full sets
Gholdengo @ Wise Glasses
Good as Gold · Timid · 2 HP / 32 SpA / 32 Spe
  • Make It Rain
  • Shadow Ball
  • Focus Blast
  • Recover
Annihilape @ Life Orb
Defiant · Adamant · 2 HP / 32 Atk / 32 Spe
  • Rage Fist
  • Shadow Claw
  • Bulk Up
  • Drain Punch
Incineroar @ Sitrus Berry
Intimidate · Impish · 32 HP / 32 Def / 2 SpD
  • Fake Out
  • Parting Shot
  • Will-O-Wisp
  • Flare Blitz
Archaludon @ Leftovers
Stamina · Impish · 32 HP / 32 Def / 2 SpD
  • Body Slam
  • Electro Shot
  • Iron Defense
  • Rest
Dragapult @ Choice Scarf
Infiltrator · Timid · 2 HP / 32 SpA / 32 Spe
  • U-turn
  • Shadow Ball
  • Draco Meteor
  • Will-O-Wisp
Kingambit @ Black Glasses
Supreme Overlord · Adamant · 32 HP / 32 Atk / 2 Def
  • Kowtow Cleave
  • Sucker Punch
  • Iron Head
  • Swords Dance
Rejected sets (3)

Attempt 1

  • No <team>...</team> block found (the answer was cut off at the token budget).

Attempt 2

  • No <team>...</team> block found.

Attempt 3

  • Archaludon can't learn Recover.
draft 3 of 3six picked on attempt 1sets legal on attempt 3

"Incineroar and Grimmsnarl provide critical support through Fake Out, Intimidate, Parting Shot, and Prankster-boosted screens and Fake Tears to enable the offensive core. Torkoal sets sun with Drought to power up Eruptions and Incineroar's Fire moves, while Gholdengo, Kingambit, and Dragapult offer diverse special, physical, and fast special pressure respectively, with items like Light Clay, Heat Rock, and Choice Scarf maximizing their effectiveness."

Full sets
Incineroar @ Sitrus Berry
Intimidate · Impish · 32 HP / 32 Def / 2 SpD
  • Fake Out
  • Parting Shot
  • Flare Blitz
  • Protect
Gholdengo @ Life Orb
Good as Gold · Timid · 2 HP / 32 SpA / 32 Spe
  • Make It Rain
  • Shadow Ball
  • Focus Blast
  • Protect
Kingambit @ Black Glasses
Supreme Overlord · Adamant · 32 HP / 32 Atk / 2 Def
  • Kowtow Cleave
  • Sucker Punch
  • Iron Head
  • Protect
Torkoal @ Heat Rock
Drought · Modest · 32 HP / 32 SpA / 2 SpD
  • Eruption
  • Earth Power
  • Stealth Rock
  • Protect
Grimmsnarl @ Light Clay
Prankster · Impish · 32 HP / 32 Def / 2 SpD
  • Reflect
  • Light Screen
  • Fake Tears
  • Spirit Break
Dragapult @ Choice Scarf
Infiltrator · Timid · 2 HP / 32 SpA / 32 Spe
  • Draco Meteor
  • Shadow Ball
  • Fire Blast
  • U-turn
Rejected sets (2)

Attempt 1

  • No <team>...</team> block found (the answer was cut off at the token budget).

Attempt 2

  • Grimmsnarl can't learn Thunder Wave.

Matches

No rated matches yet.