Free
No credit card required
$0forever
- 36 Ink daily
- 3 curated fiction-trained models
- Unlimited stories & branches
"AI Slop" - Wort des Jahres 2025Man erkennt es sofort beim Lesen.
Man erkennt es sofort beim Lesen.
Wir haben das Gegenmittel gebaut.
"AI Slop" wurde nicht ohne Grund zum Schlagwort. Dieselben flachen Formulierungen, dieselben wiederverwerteten Muster, dieselbe Prosa, die klingt, als hätte ein Chatbot ein Kostüm angezogen. Fast jedes KI-Schreibtool erzeugt das, weil fast alle auf ähnliche Allzweckmodelle setzen. Wir bringen Open-Weight-Modelle in Ihren Editor, die von der Fiktions-Community selbst auf kreatives Schreiben hin getestet wurden. Den Unterschied merkt man ab dem ersten Absatz.
Filtered from every generation
tapestrysent shivers down her spinedance of shadow and lightsymphony ofpulse quickeneda testament toher breath hitchedlittle did she knowacross the crowded roomfelt the weight ofelectric currentthe world around themtime stood stilla mixture ofcouldn't help but noticeher heart racedseemed toa labyrinth ofunspoken wordsthe air was thick with
15+ kuratierte ModelleRotierender KatalogKostenloser Einstieg
Das KI-Schreibproblem
Sie wissen längst, wie KI-Prosa klingt.
Sie haben sie generiert, sich daran gestört und versucht, sie in lesbare Form zu bringen - bis Sie aufgegeben haben.
Das sind keine kreativen Entscheidungen, sondern statistische Nebenwirkungen eines Allzweckmodells, das auf Internettexten, Code und Chatprotokollen trainiert wurde. Im Fiktionskontext ist das wahrscheinlichste nächste Wort oft das generischste. GPT-4 und Claude können gut zusammenfassen - aber selten eine Szene schreiben, die man ein zweites Mal lesen möchte.
Die besten Modelle werden nicht von Schreib-Apps gebaut, sondern von Schreibenden ausgesiebt.
Große KI-Schreibplattformen trainieren eigene Modelle oder verpacken ein Allzweckmodell in einer schicken Oberfläche. Bewertet wird mit Benchmarks und Demos - nicht damit, ob ein Modell einen 90.000-Wörter-Roman trägt.
Die Open-Source-Community testet anders. Autorinnen, Autoren und Forschende schreiben tatsächlich mit diesen Modellen, veröffentlichen damit und lesen das Ergebnis wie Schreibende, nicht wie Produktmanager.
How it works
Four moves from blank page to published chapter.
No wall of instructions. No tutorials. Each step is the one Unsloppy handles differently than every other AI writing app.
Pick a model
Twenty curated open-weight models, hand-tested for fiction. Not one giant LLM pretending to be everything.
Draft your scene
A real manuscript editor with a branching chapter tree. Your narrative is a native data type, not a Google Doc hack.
Refine with the agent
The agent already knows your lorebook, characters, and POV. No re-pasting context every turn.
Export for KDP
One-click manuscript export to KDP-ready .docx and .epub. Write here, publish anywhere.
So funktioniert es wirklich
Drei Bausteine. Alle auf Fiktion geschärft.
Ein Editor für Manuskripte, ein kuratierter Modellkatalog und ein Backend, das den Text redigiert, bevor Sie ihn lesen.
Lorebook
A living reference for your world — characters, locations, magic systems, faction politics, timeline events.
A living reference for your world — characters, locations, magic systems, faction politics, timeline events. The editor pulls the relevant entries into context as you write, so the model stays inside your world without you re-pasting it into every prompt.
Character engine
Give each character motivation, speech patterns, and contradictions.
Give each character motivation, speech patterns, and contradictions. The model remembers from chapter one through chapter forty, so when the protagonist's ex walks into the bar in the third act, the dialogue actually lands.
Rewrite tools
Highlight a paragraph.
Highlight a paragraph. Tell the editor to tighten the dialogue, cut the adverbs, raise the stakes. The rewrite comes back in your voice, not the model's.
Wir bauen keinen ChatGPT-Mantel.
MerkmalDie teuren Schreib-AppsUnsloppy
Models under the hoodOne proprietary model plus a menu of frontier APIs (Claude, GPT). The big apps all draw from the same two or three labs.Twenty-plus open-source fine-tunes from the fiction community. Different strengths — prose polish, tension, dialogue, worldbuilding — picked per scene, not per subscription tier.
Who curates themA product team choosing by benchmark score and vendor availability.A working fiction writer who tests each model the hard way — by drafting a chapter in it and seeing whether he keeps it.
What the output looks likeRaw model output. You edit out the shivers-down-the-spine and the rule of three yourself.Every generation passes through an AI-ism detector (~200 banned phrases with negative logit bias), a structural sanitizer, and a margin-suggestion pass — before you see it.
Model refresh cadenceWhatever the vendor ships. A new Claude drops in the spring; it arrives in your app in the fall.The roster rotates with the community. When a better fine-tune releases, we test it and add it. The catalog stays current without you moving.
Creative latitudeContent filters tuned for a B2B SaaS audience. Tension, violence, and sex get sanded down.Your story, your rules. We block illegal content. The rest is yours to write.
The workspaceA chat panel bolted to a document. A story bible for world consistency, but the manuscript tends to live in another tool.A full manuscript editor: lorebook, character engine, chapter tree, rewrite tools, and one-click KDP-compliant export. The book lives here.
What it costs$19 to $59 a month depending on tier, with word credits metered per generation.$19.99/mo Pro or $39.99/mo Premium, with 10,800 or 21,600 Ink dropped on signup and a daily refill. Flat pricing, no surprise meters.
Demnächst
Agent-Modus. Sagen Sie, was Sie wollen. Beobachten Sie, wie es schreibt.
Genre, Ton und Umfang festlegen - der Agent erstellt eine Struktur, schreibt Kapitel und prüft seine eigene Arbeit gegen Lorebook und Figurenkarten. Sie geben Richtung und Freigabe, das System übernimmt Volumen und Konsistenz.
01
AI-ism detector
A banned-phrase list baked into generation with negative logit bias. When "tapestry," "symphony," "dance of," "sent shivers," and two hundred of their siblings try to appear, they get pushed back down the token distribution before they ever reach your page.
02
Structural sanitizer
A post-generation pass that flattens character-card bleed-through, prompt leaks, unsolicited bullet points, runaway em-dashes, summary wrap-ups, and rule-of-three cadences — before they hit your manuscript.
03
Margin suggestions
An agentic pass reads your draft and returns zero to three targeted notes — pacing, sensory gap, flat dialogue, lost tension — in the margin, the way a developmental editor would.
“I spent a few years inside two of the biggest AI writing apps — first as an employee, then as a contractor. Good people. Real constraints. But the product was optimizing for retention, not prose. So I left to build the tool I'd actually want to finish my next novel on.”
Founder · data scientist · working novelist
Modellkatalog
Nicht ein Modell. Eine kuratierte Mannschaft.
Jedes Modell im Katalog wurde in der Community für kreatives Schreiben erprobt. Wir kennzeichnen die jeweiligen Stärken, damit Sie je nach Aufgabe das passende Werkzeug wählen können.
Free-Tier — 36 Ink täglich, für immer
Gemma 3 12B
— The Brainstormer
Rapid-fire outlining and plot beats. A fresh, eager tone that helps when you're staring at a blank page.
Recommended
Mistral Small Creative
— The Plot Pusher
Fast and direct. When a scene is stalling, this model pushes the narrative forward without descriptive overhead.
Qwen 3.5 9B
— The Continuity Checker
A pocket assistant that excels at referencing your lorebook, checking consistency, and keeping quick drafts on track.
Pro — 19,99 USD/Monat (14,99 USD/Monat jährlich)
New · Recommended
Gemma 4 31B
— The Prodigy
Exceptional prose fluency, strong dialogue, and one of the most talked-about new releases in open-source creative writing.
Qwen 3.5 27B
— The Dialogue Master
Nuanced, human-sounding dialogue and steady pacing. Excels at keeping character voice consistent across long sessions.
Cydonia 24B
— The Action Director
High-tension scenes, suspense, and dark fiction. This model drives conflict and pacing naturally.
Big Tiger Gemma 27B
— The Ruthless Editor
Cuts down on flowery AI phrasing and focuses on show-don't-tell prose.
Command R 35B
— The Ensemble Weaver
Complex scenes with multiple characters. Tracks intertwining plotlines and rarely confuses who is doing what in a crowded room.
Premium — 39,99 USD/Monat (29,99 USD/Monat jährlich)
Prose Gold
Magnum 72B
— The Wordsmith
Gorgeous, literary prose with minimal prompting. The community gold standard for raw output quality.
Euryale 70B
— The Emotional Core
Deep, resonant storytelling. Reliably captures character psychology, uncovers subtext, and maintains pacing during emotional chapters.
Qwen 3.5 122B MoE
— The Epic Novelist
Massive context window for sprawling series. Juggles numerous subplots without losing the narrative thread.
Recommended
Hermes 4 70B
— The Worldbuilder
Highly analytical and structured. Best for hard sci-fi, complex mysteries, or intricate fantasy systems with cause-and-effect plotting.
GLM-4.6 357B MoE
— The Unbound Architect
The heavyweight. Manages massive, intricate worlds with highly permissive creative guardrails.
Pricing
Flat pricing. Never a surprise meter.
Every plan includes the full writing studio. Pick your tier based on monthly writing velocity, not feature gating.
Pro
For working writers
$19.99/month
or $14.99/mo billed annually
- 360 Ink daily
- Full context window
- Economic prose models
- EPUB & PDF export
Premium
For power users
$39.99/month
or $29.99/mo billed annually
- 720 Ink daily
- Full catalog of curated models
- All export formats
Open beta · free tier
Ihr nächstes Buch wartet schon.
Hören Sie auf, Prompts in ein Chatfenster zu werfen und das Ergebnis später in Word zu reparieren. Öffnen Sie einen echten Editor, wählen Sie ein Modell, dem die Fiktions-Community vertraut, und legen Sie los.