PvE boss raids, 2v2 team battles, wave survival, and co-op creature combat. Both real-time simultaneous and drop-in/drop-out. Short and long sessions.
1-4 players take turns flicking creature marbles at a boss in a shared portrait arena. Hitting an ally triggers their Bump Combo ability (lasers, explosions, homing missiles). Each creature charges a Strike Shot special. Boss has moving weak points, countdown attacks, and hazard zones.
Monster Strike ($11.4B lifetime). Portrait. Mobile-first. Co-op was designed as the primary mode -- local co-op bonuses, shared continues, social-first design. Created by Yoshiki Okamoto (Street Fighter II creator).
Each player brings one creature to a 4-creature team with shared HP pool. Players take turns flicking (Player 1, Player 2, Player 3, Player 4, repeat). Hitting allies triggers THEIR abilities, so positioning matters cooperatively. Boss countdown attacks create shared urgency. Originally required local proximity -- co-op was the social hook that drove word-of-mouth growth.
The highest-grossing mobile game mechanic ever, and co-op IS the core mode. Turn-based flicking is naturally 1-4 player. Each player's creature contribution is visible and meaningful. Bump Combos create genuine cooperation (you position YOUR creature to be hit by your ALLY's flick). Scales from solo (you control all 4) to 4-player seamlessly. Static marble sprites + VFX explosions = minimal animation. Portrait from day one. Abilities are the entire combat system.
Never shipped PvP at scale. Team-vs-team would need design work. Session length is 3-10 minutes per quest, on the longer side for mobile.
Multiple players bring creature teams to fight a shared boss. Each player matches gems on their own board to charge their creatures' abilities. Damage combines against the boss. Alliance Wars pit guilds against guilds in coordinated attacks.
Empires & Puzzles ($500M+, Alliance Wars are the endgame co-op), Marvel Puzzle Quest (alliance events), Puzzles & Survival (co-op PvE). All portrait.
Alliance Wars in E&P: each alliance member gets a limited number of attacks against the opposing alliance's defense teams. Coordination matters -- who attacks which defense, in what order, which elements counter which defenders. Boss events: all alliance members contribute damage to a shared boss HP pool. Your individual match-3 skill contributes to collective progress.
Co-op is more "parallel solo with shared goal" than genuine simultaneous interaction. You never see your ally's board in real time. Alliance Wars add strategic coordination but the moment-to-moment play is individual. Still, this is the commercially proven co-op model for mobile creature battlers.
Two players share one side of a portrait arena. Both deploy creature cards simultaneously using separate elixir bars. Troops auto-path toward the enemy team's towers. Coordination emerges naturally: one player tanks, the other supports.
Clash Royale 2v2 ($4.9B total game revenue). Portrait. Added in 2017, became a fan-favorite mode. 4 players, 1 arena, simultaneous play, separate decks and elixir.
Both teammates see the full arena and play simultaneously. No turn-taking. Coordination via emotes and card hovering (show where you intend to deploy). Natural role specialization: one player runs heavy beatdown deck, partner runs support/spells. Friendly fire isn't possible but wasted elixir on overlapping plays is. King Tower has 2 cannons and extra HP in 2v2. No trophies at risk -- it's the "play with friends for fun" mode.
Animated sprites needed (idle, walk, attack per unit). Real-time only -- no async possible. But the 2v2 format is the gold standard for "team up with a friend" on mobile in portrait. Every creature card has unique abilities. Genuinely cooperative -- you see your ally's plays and react in real time.
2-4 players share a map. Each player owns a section and places towers/creatures to defend against waves. Shared lives pool. Income split between players. Boss rounds create shared urgency.
Bloons TD 6 co-op (2-4 players, unlocked at level 20). Map can be divided into sections or shared. Each player has own hero + tier 5 upgrades. Cash from pops is split. Players can send money to each other.
Genuine cooperation: one player handles early-game economy (farms), another handles camo detection, another handles MOAB killing. Natural specialization without forced roles. Communication through resource sharing and tower placement. If someone disconnects, their section opens up and their towers transfer to remaining players. Scales from 2 to 4 gracefully.
BTD6 is portrait on mobile but was designed for landscape/tablet. The co-op map division can feel cramped in portrait on phone screens. Tower defense is placement-heavy, not creature-battler-native. Sessions run 10-30+ minutes. PvP variant (Bloons TD Battles) exists but as a separate game, not an integrated mode.
Guild-based cooperative assaults on shared objectives. Each player brings their creature team to attack a collective target. Results compound across the alliance.
Fire Emblem Heroes ($1.2B+) -- Aether Raids, Mjolnir's Strike, Summoner Duels. Summoners War ($2B+) -- Guild Wars, Siege Battles, Labyrinth (4-player co-op dungeon). Portrait for FEH, landscape for SW.
Summoners War Labyrinth is the strongest example: 4 guild members enter a shared dungeon grid. Each member clears different rooms based on their creature roster strength. Boss rooms require multiple members' attacks to defeat. Coordination on who tackles which room creates genuine strategic cooperation. FEH's model is lighter -- guild members contribute scores to shared pools.
Co-op is usually asynchronous coordination (each player attacks separately, results combine) rather than simultaneous play. The Labyrinth model is closer to genuine co-op but still turn-based individual battles. More animation than match-3 (chibi sprites + poses). But the creature ability depth is unmatched.
Multiple players attack a shared boss creature simultaneously. Each player taps to attack and swipe to dodge in real time. Boss has a massive HP pool tuned for group damage. Timer creates urgency.
Pokemon GO raids ($6B+ total game, raids are the primary co-op driver). Up to 20 players per raid. Remote raid passes expanded access. Portrait.
Everyone attacks simultaneously in real time. Boss HP is shared -- all damage combines. No turn-taking. Each player picks a team of 6 creatures with type advantages. Mega-evolved creatures provide attack bonuses to all allies of the same type. Friendship level increases damage bonus. Genuinely simultaneous -- you see other players' creatures on screen fighting alongside yours.
The actual combat mechanic is simple to the point of shallow (tap to attack, swipe to dodge). Cooperation is mainly "everyone shows up and damages the boss" without deep interaction between players. 3D animated creatures. The social magic is in the gathering, not the gameplay. Hard to replicate without the AR/location layer.
2 players build a shared team from a combined shop. Battles resolve automatically. Each player contributes creatures and positioning decisions. Team synergies emerge from combining two players' collections.
Super Auto Pets (emoji-art creatures, simple auto-battler, free-to-play hit). TFT Double Up mode (2-player co-op, shared health, can send units to partner). Storybook Brawl (now defunct but had 2-player co-op).
TFT Double Up is the best model: two players share a health pool. During shop phase, you can send creatures to your partner. You share augment choices. Your partner fights a different opponent each round, so you're not literally on the same board, but your fates are linked. If one partner loses, the other takes damage too. Creates genuine "help your partner's build" cooperation.
All auto-battlers are landscape. Portrait auto-battler doesn't exist at scale. The horizontal team row is the core UX problem. Sessions run 20-30 minutes (too long). But the co-op quality is excellent -- sending creatures to your partner is a uniquely satisfying cooperative interaction, and it maps perfectly to "I generated this creature with AI, let me give it to you."
1-4 players survive escalating waves of enemies. Auto-attack, player controls movement only. Level up by collecting XP gems, choose abilities from random selections. 15-30 minute runs.
Vampire Survivors ($100M+, added co-op for up to 4). Brotato (solo). Survivor.io (mobile, portrait, solo). Magic Survival (mobile). None of the mobile versions have co-op in portrait.
No portrait co-op survivor game exists. Co-op in Vampire Survivors is "same screen, different characters" without deep interaction. Creature sprites need movement animation. Sessions too long for mobile burst play. But the ability/build system is extremely deep and the genre is hot.
Each player has different controls and information. Communication required to coordinate. One player might see what the boss is about to do; another player has the ability to counter it.
Spaceteam (mobile, 2-8 players, portrait, shout instructions at each other). Keep Talking and Nobody Explodes (one player sees the bomb, others have the manual). Overcooked (different kitchen roles).
The most genuinely cooperative mechanic possible -- you literally cannot succeed alone because you don't have all the information or controls. Spaceteam gives each player different panels with different controls, and instructions appear on random players' screens. You must communicate to survive. This is "cooperation as the mechanic" rather than "cooperation bolted onto a solo mechanic."
No creature battler uses this model. Creature abilities don't map naturally. Requires voice chat or physical proximity to be fun. Doesn't scale to solo play. But as a SUB-MECHANIC within a boss fight (one player weakens the boss's shield, another deals damage), this pattern creates the strongest cooperative feeling of any mechanic in this catalog.
3v3 team battles with unique characters. Each brawler has basic attack, super ability, gadgets, and star powers. Multiple game modes (Gem Grab, Brawl Ball, Heist) provide varied co-op objectives.
Brawl Stars ($2.5B+). Portrait. 70+ brawlers with unique kits. 3-minute matches. Supercell's fastest-growing title.
Heavy animation requirements -- every brawler has walk cycles, attack animations, super animations, death animations. This is a twin-stick shooter, not a simple creature battler. The skill floor is higher than tap/match/flick mechanics. Real-time only. But the co-op quality is among the best on mobile and the revenue proves the market.
The unique opportunity for Bobium is that creature GENERATION is the social act. You don't just bring your creature to the fight -- you made it. Sharing a creature you generated with a friend is inherently more meaningful than sharing a gacha pull. The auto-battler "send a unit to your partner" mechanic, combined with Monster Strike's "hitting your ally triggers their ability," creates a co-op loop where the creatures you generate directly power your friend's gameplay. This is something no existing game does.
Very few portrait mobile games have genuine real-time co-op. Monster Strike is the standout exception. Most "co-op" in portrait games is async alliance coordination (E&P wars, SW guild raids). The games with great real-time co-op (BTD6, Brawl Stars, Clash Royale 2v2) are portrait but were often designed landscape-first.
A portrait-native, real-time co-op creature battler with turn-based flicking (Monster Strike model) or simultaneous card placement (Clash Royale 2v2 model) would occupy nearly empty design space. Add AI-generated creatures that players gift to each other, and you have a social loop no existing game replicates.
Quick co-op (sub-90s): Clash Royale 2v2 pushes are fast but full matches are 3 min. No proven sub-90s co-op exists on mobile. Medium co-op (2-5 min): Monster Strike quests, Pokemon GO raids. Long co-op (10-30 min): BTD6, TFT Double Up. The sweet spot for mobile co-op appears to be 3-5 minutes based on the highest-revenue examples.