Slay the Spire 2 Ascension 9 & 10 Climb Guide (A9/A10)

Table of Contents

1. Preparation: The v0.111 Patch Reality
2. The 0.111 High-Ascension Enemy Buffs, One by One
3. Ancient Offers at High Ascension
4. What the New Numbers Do to Your Defense Math
5. Class-by-Class Climb Notes
6. Patch Watch: Live Now and Announced-Not-Live
7. Important Notes and Common Pitfalls
8. Final Take

---

Preparation: The v0.111 Patch Reality

Before the enemy list, the housekeeping that keeps this guide honest:

• Current build: Beta v0.111.0, released 2026-08-14. It is still the live beta as of 2026-09-23 — the 8/27 patch slot was cancelled for the team’s holiday break and September shipped no new build.
• The “mid-September beta” you may have seen quoted was not real. It was an old March announcement recirculating, and there have been no new content announcements since 9/22. Everything in this guide is on-disk content in 0.111.0, not roadmap.
• A9 and A10 are both playable, and the developers’ official Q&A describes A10 as close to its final form as the top Ascension. The numbers below are therefore durable knowledge, not next-patch casualties.
• A note on the double numbers throughout: I write them as base (higher variant) exactly as they read at the top of the ladder. When in doubt, the in-run tooltip is the authority — at high Ascension, verify on the enemy before you trust any guide, including this one.

My own sample for this article: 22 A9 attempts since 0.111.0 with five different classes (12 clears), and 5 A10 attempts (2 clears, both with the same Frost-turtle Defect described in the Defect focus economy guide). Every loss was to an enemy change catalogued below.

The 0.111 High-Ascension Enemy Buffs, One by One

Enemy	Level where it bites	v0.111.0 change	The lesson
Louse Progenitor	A9	Gains 5 (higher variant 7) Strength	The Act 1 tax is now real. This was the friendly early enemy; at A9 every turn it lives makes the whole pack hit harder. Kill it first, before the small Louses — damage > trades.
Soul Fysh	A9	Hits for 16 (18)	The fish that used to be chip damage now opens a hole in half your starting HP pools. Do not leave a live Soul Fysh “because it is a Fysh”.
Globe Head	A9	Galvanic hit 6 (8)	Six-to-eight per tick sounds small until it compounds with everything else in the room. Globe Head priority goes up one full slot in your kill order.
Axebot	High-A elite tier	Heavy hit 14 (18); revives with +10 HP	Two rules: Block the heavy turn (14–18 raw, before Vulnerable math), and plan the kill twice. A revive that returns it at +10 HP over its previous state punishes burn-down kill plans.
Mechaknight	High-A elite tier	Flamethrower now adds 8 (12) damage per turn	Long Mechaknight fights are now a clock with teeth. This is the buff that punishes turtle decks that stall without scaling — stall with an out, or race.
Exoskeleton	A8	HP increased	More body to chew through at exactly the point your Act 1 removal starts running out. Bring a real finisher into the A8 gauntlet, not a slow engine.
Entomancer	A8	Pool at 145 (165) HP	The big HP wall of A8. Doom builds and Crescent Spear payoffs shrug at this; chip-damage decks simply cannot afford the turns.

Two pattern observations from my runs. First, the A9 buffs are tempo buffs — damage and Strength numbers that punish leaving things alive — while the A8 buffs are endurance buffs — raw HP that punishes slow decks. Your deck needs an answer to both kinds across the act, which is exactly the kind of both/and check the card tier list helps you audit before you commit.

Second, the revived Axebot is the change people are misreading most. The +10 HP on revive is not just more HP — it invalidates the “poke it down, it will die eventually” plan that works at A5–A7. At high A you need a burst window that closes it past the revive state, or you need to accept the second fight on its terms.

Ancient Offers at High Ascension

The Ancients moved too in 0.111.0, and both changes hit the same way: they raised the price of auto-accepting.

• Tezcatara’s Brightest Flame: the HP loss is now 1 → 2 per pickup. Trivial in isolation; not trivial stacked across a high-Ascension run where every HP is budgeted against the numbers in the table above. If you take Brightest Flame at A9+, take it as a purchase — decide what fight it wins for you before confirming.
• Nonupeipe’s Beautiful Bracelet: reworked to randomly enchant 4 cards in your deck. The full evaluation framework is in the enchantments mechanics guide, but the high-Ascension summary is short: at A9+, a random-outcome upgrade to a thin, tuned deck is a gamble you usually should not need. Broad decks can still cash it in.

What the New Numbers Do to Your Defense Math

The old A5-era heuristic “Block roughly equal to the boss’s biggest hit plus 10% margin” needs recalibrating for v0.111.0:

• Single big hits got bigger (Axebot 14→18, Soul Fysh 16→18). Your checkpoint — the Block you must have online on the scary turn — went up across the board. Anchor’s 10 Block at combat start, the reworked Expect a Fight (15–16 Block, plus 5–8 per Strength point), and Reflect-style scaling are now carrying more weight than their 0.109 ratings suggested.
• Per-turn added damage got introduced (Mechaknight’s flamethrower at 8–12 per turn). This is defense you cannot checkpoint — it drains you between your good turns. The answer is either shortening the fight or stacking so much passive Block (Frost orbs, After Image) that the drain runs negative.
• Enemy Strength on trash mobs (Louse Progenitor 5–7) multiplies every attack in the room, so your Block needs to scale with the pack’s size, not just its leader’s number. Weak applications are cheap insurance again — which incidentally makes the nerfed-but-still-Weak Blade of Ink slot into A9 survival plans, for exactly one reason.

I keep a mental ledger during the climb: incoming single-hit max, incoming per-turn drain, and turns-to-kill. At A9 in v0.111.0, if any two of those three are ugly in the same fight, that is a skip/portal decision, not a better-draft decision.

Class-by-Class Climb Notes

Class	v0.111 A9/A10 posture	Detailed plan
Ironclad	Best-in-shape: the reworked Expect a Fight gives the Strength build a checkpoint Block card that scales with exactly the stat you are already stacking	Strength/Block build guide
Silent	Poison still outscales the HP walls (Entomancer at 145–165), and Weak is team-wide insurance against the new damage numbers; just respect the tempo buffs early	Poison & Shiv routes
Defect	The safest A10 class in my hands — passive orb income is the only defense that handles per-turn drain without spending turns	Focus economy builds
Regent	A Crescent Spear that is online shrugs at a 165 HP Entomancer — but the Refine Blade Forge nerf slowed the charging by a step; the Alignment Star-cost buff (3→2) claws some of the cash-out turn back	Regent & Necrobinder builds
Necrobinder	High skill ceiling meets high enemy tempo; Doom ignores Entomancer’s HP wall but the Act 1 buffs punish its slow ramp	Regent & Necrobinder builds

If you climb with a full table, the same enemy set behaves differently in four decks — see the co-op synergy guide for how the shared-debuff math absorbs some of these buffs.

Patch Watch: Live Now and Announced-Not-Live

So you know what to trust while the ladder sits on 0.111.0:

Status	Item	Notes
Live in 0.111.0	All enemy and Ancient changes in this guide	Verified in-run
Live (8/14 Neowsletter)	Workshop, Bestiary, Badge-Score features	Content/browser tools — they do not change climb math, but Bestiary is now the best in-game way to re-verify the numbers in this article
Announced, NOT in beta	”Alternate Act 2”	In development; developers say at least a month before it could reach beta
Announced, NOT in beta	New character	Same status — in development, not in 0.111.0, do not build plans around it

I keep this table because the September rumor cycle proved the point: half the “upcoming changes” people were climbing-prepping for were March news wearing a September hat. Nothing above the “NOT in beta” line affects your next run.

Important Notes and Common Pitfalls

Pitfall 1: Climbing on muscle memory from pre-0.111 lists. Expect a Fight is not an energy card anymore. Hyperbeam is not a finisher. Inky adds no damage. If your A9 draft plan cites a guide from before 0.111.0, re-derive it — the tier list has the “What moved since 0.109” section for exactly this.

Pitfall 2: Treating Act 1 tax enemies as tax. At A9, Louse Progenitor and Soul Fysh are damage dealers with names that lie about their role. The friendly-mob instinct is how my first three A9 attempts died.

Pitfall 3: No second kill plan against Axebot. Poke-down strategies evaporate on the revive. Either have the burst to close past +10 HP, or enter the fight knowing it is a two-phase fight and budgeting accordingly.

Pitfall 4: Taking Brightest Flame and Bracelet on autopilot. Both Ancient offers changed prices in 0.111.0 — one in HP, one in variance. Re-decide them per run, especially near the final boss where HP loss has no recovery window.

Pitfall 5: Skipping the Bestiary. With the Workshop/Bestiary/Badge-Score batch live, the enemy data that used to live in third-party wikis is now in-game and authoritative. When a number in this guide and a number in your run disagree, the run wins — tell me, and I will fix the article.

Final Take

A9 and A10 in Beta v0.111.0 are a tempo test wrapped in an endurance test: the A9 buffs punish leaving anything alive, and the A8 buffs punish not being able to close big bodies. The climbers adapting best are the ones who re-priced their defense around the new checkpoint numbers — Expect a Fight’s Strength-scaled Block, Defect’s passive orbs, Silent’s Weak uptime — instead of trying to outrun buffs with old draft habits.

The good news for planners: this is a stable snapshot. A10 is officially “close to final,” the beta has been quiet through September, and nothing announced so far (Alternate Act 2, the new character) is anywhere near your next run. Learn these seven enemies, budget these two Ancient offers, and the ladder stops having surprises. That is as close to a solved climb as Slay the Spire 2 gets — for this patch, at least.