Larian Studios is enjoying a victory lap after Baldur’s Gate 3 scooped up nearly every 2023 Game of the Year award, but the Belgian RPG shop took a 27-year detour through cancelled projects, bankrupt publishers, and improvised side hustles to get here. Their story is less about sudden success and more about relentless iteration on an idea: build systemic, reactive RPGs that feel like a tabletop session with friends.

Understanding how Larian grew from a dozen people in Ghent to a 450-person multi-studio developer explains why its games feel different. This deep dive traces the company’s origins, the survival tricks that kept it afloat, and the design culture that now shapes a sizable chunk of the western RPG market.

From Ghent experiments to Divine Divinity

Swen Vincke founded Larian in 1996 in Ghent, Belgium, with the dream of building immersive sims at a time when most European studios chased licensed platformers. Early prototypes like The L.E.D. Wars (1997) and the never-released Project C taught the team how to ship with comically little money. The breakthrough came with 2002’s Divine Divinity, an isometric RPG that mixed Diablo-style combat with the playful systemic design Vincke loved in Ultima VII. Sold through CDV, it moved more than a million copies over time and gave Larian just enough credibility to attempt a sequel.

Near-bankruptcy and lessons learned

Success didn’t equate to stability. Beyond Divinity (2004) launched in a broken state because Larian’s publishing deal paid on milestones rather than quality, and the studio nearly folded. To keep the lights on, Vincke spun up side businesses, from educational games for Belgian broadcaster VRT to outsourced projects like KetnetKick. Those gigs funded passion experiments such as the cancelled Game One MMO and the troubled Divinity II (2009), which shipped incomplete and later required the Dragon Knight Saga re-release to fix.

The lesson was clear: relying on third-party milestones left Larian perpetually broke. Vincke publicly vowed to self-publish, even if that meant fewer marketing dollars. That stubbornness set the stage for the studio’s next act.

Rebooting with Kickstarter and Divinity: Original Sin

By 2012, Larian bet the company on a modern co-op RPG that embraced systemic chaos. With publishers skeptical, the team went directly to players. The Divinity: Original Sin Kickstarter (2013) raised roughly $1 million, but the more important resource was the community feedback loop. Backers stress-tested every spell combo, nudging Larian toward the sandbox combat that now defines the brand. When the game launched in 2014, reviews praised its freedom, and Vincke immediately reinvested profits into a sequel.

Divinity: Original Sin 2 (2017) doubled down on everything fans loved—drop-in co-op, environmental spell reactions, full voice acting, and a “Game Master Mode” that turned the engine into a tabletop toolkit. It sold over two million copies within its first year, proving that high-budget CRPGs could thrive without microtransactions or live-service hooks.

Winning the Baldur’s Gate 3 mandate

Hasbro and Wizards of the Coast took notice. In 2019, they invited Larian to pitch for Baldur’s Gate 3, entrusting the Dungeons & Dragons license to a studio outside the United States for the first time. Larian agreed on the condition that it retained creative control and could ship through early access. That decision defined the project: a public alpha arrived in October 2020, letting players poke holes in every encounter and romance path for nearly three years.

The 1.0 release in August 2023 became a cultural moment. It topped Steam’s charts, maintained a 96 on Metacritic, and swept the 2023–2024 awards circuit. Larian’s approach—long early access, constant hotfixes, and surprise content updates like Patch 5’s epilogue—turned an “old-school” CRPG into a mainstream phenomenon.

How Larian actually builds its worlds

Larian now operates seven studios (Ghent, Quebec City, Dublin, Guildford, Barcelona, Kuala Lumpur, and Warsaw) to keep production follow-the-sun. Each location owns specific pipelines—Ghent leads narrative and systems, Quebec handles cinematics, Kuala Lumpur focuses on art outsourcing, and Dublin manages publishing and community. Everyone feeds into a single production philosophy:

• Systems first: Designers prototype spells, surfaces, and AI routines in a shared editor; narrative teams layer quests after the toy box feels fun.
• Everything is voice-acted: After Divinity: Original Sin 2 proved fully voiced CRPGs sell, Larian invested in its own mocap stages (Ghent) and partner booths (London, Los Angeles) to record thousands of lines quickly.
• Player-driven QA: Early access is treated as a real production phase. For BG3, more than 2.5 million players joined before launch, effectively becoming an external QA department.
• No monetisation creep: Larian’s leadership repeatedly rejects battle passes or paid cosmetics, preferring single, premium SKUs plus free patches. That choice builds goodwill but requires massive up-front capital, which Vincke funds by keeping the company private and reinvesting revenue.

Business model and future bets

Larian remains privately held, with Vincke controlling the board. Instead of chasing acquisitions, the studio buys time: it owns its Ghent campus, leases long-term spaces in Canada and the UK, and maintains a war chest built from BG3 sales. The company also runs its own publishing label, so distribution on Steam, GOG, and console storefronts flows directly back into development. That independence lets it take unusual swings—like shipping a full orchestra tour or funding community toolsets—without shareholder pressure.

Post-BG3, Vincke told fans not to expect DLC or Baldur’s Gate 4 anytime soon; Larian is already working on “two very ambitious new RPGs,” widely rumored to include a fresh fantasy IP and a sci-fi project. The studio is also hiring aggressively in Warsaw for combat designers and engine programmers, hinting at even more systemic ambition.

What to watch next

Larian’s next challenge is sustaining momentum without burning out its team. Expect the studio to keep iterating on the Divinity engine, refine its cinematic pipeline, and push its internal tools closer to tabletop flexibility. If it can deliver another premium RPG without compromising that “play however you want” ethos, Larian could cement itself as the go-to house for narrative sandboxes—licensed or original.

For now, the playbook is clear: stay independent, listen to players early, and never stop teaching spells to set furniture on fire.