Guildrun Tier List

Version and evidence status

Grace, Yuuna, Karsu, Niklas, Nyx, Logan, and Reyna are placed in B Tier. No hero is forced into C Tier, while Dragomir, Funke, Hoyoung, Ming, Pimenta, Ratna, Rip, Rowan, and Zuri remain in a Testing group because the author found insufficient current evidence for a letter grade. The ranking is not intended to be permanent. Positioning, specializations, items, relics, difficulty, and team composition can move a hero above or below the displayed tier.

A strong specialization also does not prove that every version of the same hero deserves the same placement. Version and evidence status The article was evaluated against Demo version 0.5, released on July 16, 2026. Its strongest evidence comes from Leyline's official balance commentary, which identifies mode-specific high performers and explains the reasons behind buffs and nerfs. The tier letters are the article author's interpretation of those official performance statements.

Reddit reports add practical context but are not treated as controlled tests. The page recommends rechecking every placement after a balance update, specialization change, bug fix, or meaningful change in public run data because the official Steam description promises weekly updates. Ranking methodology Flexibility: Whether a hero retains value across different teams, specializations, items, relics, and mid-run pivots. Scaling: Whether the hero remains useful in SSS, Red Rift, and Endless rather than depending only on an early power spike.

How the ranking methodology works

Survivability: Whether the hero performs reliably with reasonable protection instead of requiring a perfect formation or defensive roll. Team utility: Value beyond personal damage, including shields, debuffs, control, Backup effects, and support for a shared build engine. Secondary considerations: Ease of use, item and relic dependence, positioning tolerance, pivot value, and consistency across modes. S Tier means that available evidence supports unusually broad or repeatable value.

A Tier means a hero is strong under clearer conditions. B Tier means the hero is useful but more dependent on a build, mode, or timing window. Testing means that there is too little repeatable evidence to assign a tier; it is not a rank below C. Version 0.5 editorial table Selected hero evidence Gustav receives a high-confidence provisional S placement because official patch notes described him as a top hero in SSS and Red Rift and as a hero beginning to top Endless leaderboards.

The article distinguishes this official cross-mode statement from its own decision to translate that evidence into S Tier. Tilly's placement draws on separate specialization signals. The Sharp was described as a top performer in SSS and deep Endless, while The Elegant was strong in Red Rift. The page warns that neither statement proves every Tilly build is S Tier, especially because both highlighted paths were adjusted.

How to use the tiers without overgeneralizing

Kai is treated as an A-tier anchor. Leyline said all three of his specializations were playable and strong and that he had been outperforming other Warriors before version 0.5 tuning. His broad viability supports the placement, while the subsequent scaling reductions prevent the article from treating it as permanent.

Grace is placed in B because official notes described her baseline as healthy but also showed a sharp difference between on-board and Backup value depending on specialization. Logan is placed in B because The Juggernaut has a strong late-game ceiling but a slower and more conditional start. Reyna is treated as mode-sensitive: earlier official notes supported her Endless value while indicating weaker Red Rift performance.

How the page says to use the list The article separates official facts, editorial assessments, community observations, and remaining test requirements. A hero in Testing is not assumed to be weak. It means the available reports did not consistently include the full team, difficulty, items, relics, formation, and exact game version required for a defensible comparison.

Official scope

The official Steam page was checked on 2026-08-14. It describes Guildrun as a PvE autobattler roguelike in which the player turns a simple team into a powerful force, combines heroes beyond their original class roles, and builds around specializations, items, and relics. The official page lists 25 heroes, 180 specializations, more than 300 relics, more than 100 items, and eight difficulty levels. It also describes drafting, leveling, respeccing, saving during a run, and adapting teams across randomized runs.

Weekly game updates are part of the official product description. The official Steam page does not publish an S/A/B/C/D hero tier list and does not validate the placements made by either Google source. Its official material instead establishes why rankings are conditional: hero value changes with the draft, specialization path, items, relics, formation, difficulty, and balance update. The third-party rankings above therefore remain attributed to their individual sources rather than being presented as an official Guildrun tier list.

How to verify the recommendation yourself

Identify the active patch before comparing a guide with your current game. Inspect the offered specialization, item, relic, class condition, and timing window instead of relying on a label alone. Test whether the frontline survives and whether the damage engine still works against the next boss. Record the exact difficulty, team, ranks, items, relics, and formation when a result looks unusually strong or weak.

Keep a safe pivot because random shops and rewards can prevent an exact recipe from appearing. Do not turn one creator run or one community observation into an official rule. A build is coherent when rank, class tags, timing, equipment, relics, and positioning support the same win condition. A recommendation becomes weaker when it depends on an unrecorded patch, an unknown specialization, or a perfect shop sequence.

Any unsupported number, name, reward, or release claim remains 待确认. Before combat, state which hero or effect is expected to provide survival, damage, and utility. After combat, separate a positioning failure from a rank, item, relic, or timing failure. Compare the value of an immediate upgrade with the cost of waiting for a more specific connection.

Check whether an item is on the hero that actually scales from its stats or trigger. Check whether a relic fires often enough for the current team rather than only in a perfect future setup. Treat Rush and Stall as different timing plans instead of mixing them without a clear payoff.

Prepare for the next known encounter before optimizing only for a distant Endless ceiling. When two sources disagree, keep both version and mode labels visible instead of averaging the claims. Use successful runs to form a testable hypothesis, then verify it across different shops and rewards.

Frequently asked questions

Sources used

• guildrun.app — source
• games.gg — source
• store.steampowered.com — source

Research collected August 14, 2026. Community and creator evidence is identified as non-official; unsupported details are marked 待确认.