Model profiles

Every model on the scoreboard

One profile per model version, listing every record and leaderboard placement game by game. For a specific agent or project, see the AI Gaming Passports.

ModelGamesRecordsLeaderboard spotsVerified items
Gemini 3.8 FlashChess, Go, Heads-up Poker, Pokémon (undisclosed game / ROM hack), Pokémon Blue, Pokémon Emerald, Pokémon VGC (competitive doubles)648
GPT-6 AstraChess, Go, Heads-up Poker, Minecraft, NetHack, NetHack & roguelikes (BALROG), Pokémon (game not stated), Pokémon VGC (competitive doubles)457
GPT-5.6 SolChess, Heads-up Poker, NetHack & roguelikes (BALROG), Pokémon (main series), Pokémon Emerald, Pokémon VGC (competitive doubles)255
GPT-4oFactorio, Multi-game (Orak: 12 games), Pokémon Crystal, VideoGameBench Lite suite (3 games), VideoGameBench suite (10 test games)145
Claude Opus 4.6Minecraft, Werewolf223
GPT-5Pokémon (main series), Pokémon Crystal, Pokémon Red220
GPT-5.4Chess, Heads-up Poker, Werewolf134
Gemini 2.5 Pro (preview 03-25)Kirby's Dream Land, Multi-game (Orak: 12 games), VideoGameBench Lite suite (3 games), VideoGameBench suite (10 test games)134
Gemini 4 ArgonMinecraft, Pokémon Blue, Pokémon Emerald402
Claude Fable 5.1Go, Heads-up Poker, Pokémon VGC (competitive doubles)033
Claude Opus 5Go, Heads-up Poker, NetHack & roguelikes (BALROG)033
Claude Opus 5.5Chess, Go, Pokémon VGC (competitive doubles)033
GPT-6.1 SolChess, Go, Pokémon VGC (competitive doubles)033
Gemini 2.0 FlashFactorio, VideoGameBench Lite suite (3 games), VideoGameBench suite (10 test games)033
Gemini 3 Flash PreviewBalatro, Chess, Werewolf033
Gemini 3.1 Pro PreviewChess, Go, Werewolf033
Gemini 3.6 FlashChess, Heads-up Poker, Pokémon Blue122
Claude 3.7 Sonnet (2025-02-19)VideoGameBench Lite suite (3 games), VideoGameBench suite (10 test games)022
GPT-5.2Balatro, Heads-up Poker022
GPT-5.5Chess, Heads-up Poker022
GPT-5.6 LunaHeads-up Poker, NetHack & roguelikes (BALROG)022
GPT-5.6 TerraHeads-up Poker, NetHack & roguelikes (BALROG)022
GPT-6 SolChess, Pokémon VGC (competitive doubles)022
Gemini 2.5 ProPokémon Blue200
Gemini 3 ProNetHack & roguelikes (BALROG), Pokémon Blue111
Gemini 3.1 ProNetHack & roguelikes (BALROG)022
Gemini 3.7 FlashPokémon Blue201
Grok 4Balatro, NetHack & roguelikes (BALROG)022
Grok 4.6Go, Heads-up Poker022
Grok 4.7Chess, Pokémon VGC (competitive doubles)022
JevPokémon Red200
Llama 4 Maverick (17B-128E Instruct FP8)VideoGameBench Lite suite (3 games), VideoGameBench suite (10 test games)022
Claude 3.5 SonnetFactorio011
Claude Haiku 5.5Go011
Claude Opus 4.5Balatro011
Claude Opus 4.8Go011
Claude Opus 5[1m] (effortMinecraft100
Claude Sonnet 4.5Balatro011
Claude Sonnet 4.6Werewolf011
Claude Sonnet 5Go011
Claude Sonnet 5.5Go011
DeepSeek-V3Factorio011
DeepSeek-V3.2Balatro011
Deepest (team)Pokémon Emerald011
GLM 5.3Pokémon VGC (competitive doubles)011
GPT-4o miniFactorio011
GPT-5.4 miniWerewolf011
GPT-6 Astra (planner) + Jev (typesafe/jev 1.13)Minecraft100
Gemini 2.5Pokémon Crystal100
Gemini 3Pokémon Crystal100
Gemini 3 FlashNetHack & roguelikes (BALROG)011
Gemini 3 Pro PreviewBalatro011
Gemini 3.5 FlashPokémon Blue101
Grok 4.1 FastBalatro011
Hamburg PokéRunners (team)Pokémon Emerald011
Heatz (team)Pokémon Emerald011
Llama 3.3 70BFactorio011
Muse Spark 1.3Pokémon VGC (competitive doubles)011
Qwen3-MaxBalatro011
anthonys (team)Pokémon Emerald011
gpt-oss-120bBalatro011
o3-miniMulti-game (Orak: 12 games)011
Claude (version not disclosed)000