Diffit is a free AI tool for teachers that turns a topic, link, pasted text or PDF into a reading passage, vocabulary list and comprehension questions at a chosen reading level. One source text becomes several versions, which suits a mixed-level ESL class in under a minute.
Last updated 1 September 2026.
Key takeaways
Dated reference data
The one authoritative figure set for Diffit’s tiers and limits, taken from Diffit’s own pricing page and checked on 1 September 2026. Software pricing and features change frequently, so reconfirm before any purchasing decision.
Sources. Tiers, limits and levels: Diffit’s official pricing and features page, checked 1 September 2026. CEFR level descriptions: Council of Europe, Common European Framework of Reference for Languages, checked live 1 September 2026.
How to cite this. tefl.ai, “What Is Diffit? A Practical Guide for English Teachers”, updated 1 September 2026, https://tefl.ai/what-is-diffit/. Product details sourced from Diffit’s pricing page, checked 1 September 2026.
Diffit is a tool for doing differentiation, which is a teaching approach rather than a piece of software. If the term itself is what you came for, the approach and the vocabulary around it sit in our TEFL teaching methods and classroom techniques glossary, alongside the other terms a lesson plan is built from. Two neighbouring entries worth reading with this one are the affective filter, which explains why a text pitched too high stops being absorbed at all, and our guide to adapting reading materials with AI for mixed-level classes, which covers the workflow rather than the tool.
Diffit is a web-based AI tool built for teachers that takes a piece of content, a topic you type in, a URL, a pasted excerpt, or an uploaded PDF, and turns it into classroom-ready materials at a chosen reading level. Feed it “the water cycle” or paste in a news article, and within roughly thirty seconds to a couple of minutes it returns a leveled reading passage, a short summary, a key vocabulary list with definitions, and a set of comprehension questions ranging from multiple choice to open-ended discussion prompts, according to Diffit’s own pricing and features page.
It was built originally for K-12 differentiation, the general instructional approach of adjusting content and process to fit a range of student readiness levels described in the U.S. Department of Education’s LINCS guide to differentiated instruction, adjusting the same content for students reading two or three grade levels apart in the same class. That same mechanic, one source text producing multiple reading levels, maps directly onto an ESL classroom with a mix of CEFR levels sitting in the same room, which is likely why English teachers have picked it up outside its original design brief.
The name itself is a small hint at the pitch: differentiation, shortened. The tool’s own marketing leans on the idea that any piece of content, a news story, a textbook chapter, a YouTube video transcript, can become several classroom-ready versions without a teacher retyping a word of it. For a mainstream classroom teacher juggling one curriculum across a wide spread of reading abilities, that’s the core promise. For an ESL or EFL teacher, the same promise reads slightly differently: instead of native-speaker grade bands, the practical use case becomes turning one authentic text into a graded set of versions across a CEFR-mixed class.
A teacher gives Diffit one of five input types: a topic or general subject (“photosynthesis,” “the history of jazz”), a URL to an article or YouTube video, a block of pasted text, an uploaded PDF, or a vocabulary list of up to ten words. Diffit then generates a set of outputs built around that input: an adapted reading passage at the selected level, a summary, key vocabulary with student-friendly definitions, multiple choice questions, short answer questions, open-ended discussion prompts, and in some cases a graphic organizer, based on the tool’s own product documentation and demo material.
Reading level selection is where the tool earns its usefulness for a mixed class. Rather than producing one fixed text, Diffit can generate the same underlying content at several different reading levels in the same session, so a teacher working with a spread of proficiencies can hand out three versions of the same core material without writing three separate texts by hand. Translation support extends this further: vocabulary and some content can be translated into a large number of languages, which helps newcomer or lower-level students access grade-level topics in their first language alongside the English version.
Diffit runs a freemium model. The free Basic tier gives quality, classroom-ready content, the ability to download materials as PDF or print and share them, editing and adapting of generated resources, and access to anything created in the last 90 days, per Diffit’s pricing page. For a teacher who wants a quick leveled text once or twice a week, the free tier covers the job without a subscription.
The paid Diffit for Schools tier adds exporting directly to Google Docs, Slides, Forms, and Classroom, or to Microsoft equivalents, alignment to academic standards, skills, MTSS tiers, and depth-of-knowledge levels, access to every resource the account has ever generated rather than just the last 90 days, a library of graphic organizers to attach to any passage, plus an admin dashboard, free professional development, and dedicated support. It’s priced as a flat-rate annual subscription tiered by student enrollment, aimed at whole-school or district purchasing rather than an individual teacher’s personal budget.
| Feature | Diffit (free tier) | Manual adaptation | tefl.ai AI Materials Adaptor |
|---|---|---|---|
| Input types | Topic, URL, pasted text, PDF, vocab list | Any source, fully teacher-controlled | Pasted text or existing materials |
| Leveling basis | Grade level, roughly 2nd through college | Teacher judgment, no fixed scale | CEFR bands (A1-C1) |
| Time per adapted text | Under a few minutes | 15 to 45 minutes, depending on length | Under a few minutes |
| Built for ESL specifically | No, built for K-12 differentiation generally | Depends entirely on the teacher | Yes, CEFR-native by design |
| Free tier limits | 90-day resource history, no LMS export | No limits, but no automation either | Free tool, standard usage limits apply |
| Cost | Free (Basic); paid Schools tier for exports and full history | Free (teacher’s time only) | Free |
Comparison compiled from Diffit’s official pricing page and tefl.ai product documentation, checked 1 September 2026; feature sets may change over time and should be reconfirmed on each provider’s site before purchasing decisions.
Diffit’s own materials describe it as a general K-12 differentiation tool, not an ESL-specific product, and that shows up in a few places. Its leveling scale runs on grade-level bands rather than the CEFR framework most ESL teachers plan around, so a teacher has to translate “6th grade reading level” into “roughly B1” by feel rather than getting a CEFR tag directly. It also has no built-in concept of second-language acquisition stages, code-switching support, or the kind of scaffolding an EFL-specific tool might default to, because that isn’t the population it was designed around.
None of that makes it unusable for English teaching, plenty of ESL and EFL teachers use it precisely because grade-level text simplification and CEFR-level text simplification solve a similar underlying problem: taking one source and producing several accessible versions of it. But a teacher choosing between a general differentiation tool and one built specifically for language learners should know which trade-off they’re making. A CEFR-native tool like tefl.ai’s AI Materials Adaptor starts from the framework ESL lesson planning already uses, so there’s no translation step between the tool’s output and a CEFR-based scheme of work.
From one input, Diffit typically produces a bundle of connected resources rather than a single output. Based on the tool’s own documentation and demonstration material, a single generation usually includes an adapted reading passage at the chosen level, a short summary of the same content, a key vocabulary list with definitions pitched at that level, multiple choice comprehension questions, short answer questions, open-ended discussion questions, and, depending on the topic, a graphic organizer.
For a teacher building a full reading lesson, that bundle covers most of the raw material in one pass: the text itself, the pre-teaching vocabulary, and two tiers of comprehension check, closed and open. What it doesn’t do is sequence those pieces into a lesson with timing, a warm-up, and a closing task, that structuring step is still down to the teacher, or a separate lesson-planning tool built for that purpose.
It’s worth being specific about what “adapted” means here, since the word gets used loosely across different tools. An adapted passage from Diffit keeps the same underlying facts and structure as the source, the same sequence of events in a news story, the same core explanation in a science topic, but rewrites the sentence length, vocabulary, and syntax to suit the target level. It isn’t a summary in the sense of cutting content out; it’s closer to a rewrite that keeps the same information at a more accessible reading difficulty. That distinction matters for a teacher deciding whether a shorter, simpler version of a text still covers everything the lesson’s comprehension questions need it to cover.
Manual adaptation gives a teacher total control and zero dependency on any tool, but it costs real time. Simplifying a single article into three reading levels by hand, rewriting sentence structure, swapping vocabulary, and writing fresh comprehension questions for each version, commonly takes a teacher fifteen to forty-five minutes depending on the source length and how many levels are needed. Multiply that across a five-class week and it adds up to hours that could go toward feedback, planning discussion activities, or simply going home on time.
This mirrors a wider point raised in UNESCO’s reporting on AI and the future of education: technical output alone doesn’t guarantee a good learning outcome, human judgment applied to that output still does most of the real work. An AI adaptation tool compresses that same task into a few minutes, but the trade-off is that the teacher didn’t write the text, so a review pass matters just as much here as it does with any AI-generated classroom material. A generated passage might use a word that’s technically at the right grade level but carries a connotation the teacher wouldn’t have chosen, or a comprehension question that’s phrased in a slightly odd way. The realistic workflow looks the same across most AI materials tools: generate, read it through once, adjust anything that doesn’t fit, then use it. That review step is short, usually a minute or two, but skipping it is where problems creep into a lesson.
The two tools solve a similar problem from different starting points. Diffit is a broad, grade-level differentiation tool that many English teachers have adopted because its underlying mechanic, one text becoming several reading levels, happens to suit a mixed-level ESL class even though that wasn’t its original design target. tefl.ai’s free AI Materials Adaptor starts from the other direction: built specifically around CEFR bands, so the output maps directly onto the scheme of work most TEFL and ESL courses already use, without a grade-level-to-CEFR conversion step in between.
Neither tool needs to be the only one a teacher uses. Some reach for Diffit when adapting general-interest reading content, a news article or a nonfiction topic, and reach for a CEFR-native tool when the task is leveling material that’s going to sit inside a formal course scheme with defined proficiency bands. Trying both on the same piece of source text is a fast way to see which output needs less editing for a specific class.
Diffit suits a teacher who regularly needs to turn general reading content, news articles, nonfiction topics, textbook excerpts, into multiple reading levels for one class, and who’s comfortable working in grade-level bands rather than CEFR. It’s a strong fit for content-based ESL instruction, where the text itself is about a real-world topic rather than a graded course-book passage, and for teachers supporting students across a wide proficiency spread in the same room.
It’s a weaker fit for a teacher whose planning is built entirely around CEFR descriptors and who needs every resource to slot into that framework without manual conversion, or for anyone who needs LMS export, standards alignment, or full historical access to past resources without paying for the Schools tier. For those cases, a CEFR-native adaptor or a paid Diffit for Schools subscription will fit the workflow more directly than the free Basic tier.
It’s also worth thinking about how a tool like this fits into a whole term, not just one lesson. A single leveled reading passage solves a single lesson’s problem. A teacher who wants a repeatable system, the same leveling step applied consistently across every unit for an entire term, benefits from settling on one tool and one workflow rather than switching between a general differentiation tool and a CEFR-native one depending on the week. Picking based on which framework the rest of the course already runs on, CEFR bands versus grade-level bands, is usually the deciding factor once the novelty of testing a new tool wears off.
Diffit itself is available directly at app.diffit.me, where the free Basic tier is enough to test whether its grade-level output works for a specific class before considering the paid Schools tier. For teachers who plan primarily around CEFR bands and want a leveling tool built for that framework from the ground up, tefl.ai’s free AI Materials Adaptor is worth testing on the same source text side by side, since the fastest way to find the right tool for a given course is comparing both outputs on one real lesson rather than choosing on paper.
Diffit takes a topic, URL, pasted text, or PDF and generates a leveled reading passage, vocabulary list, summary, and comprehension questions at a chosen reading level. Teachers use it to quickly create multiple versions of the same content for classes with a wide range of reading abilities.
Diffit has a free Basic tier that covers classroom-ready content generation, downloading, printing, and editing, plus access to resources created in the last 90 days. A paid Diffit for Schools tier adds LMS exports, standards alignment, and full resource history.
No, Diffit was built as a general K-12 differentiation tool and uses grade-level bands rather than the CEFR framework. Many ESL and EFL teachers use it anyway because the core mechanic, one text producing several reading levels, transfers well to a mixed-proficiency language classroom.
Diffit generates content across a wide span, roughly from 2nd grade through college level, based on the tool’s own product documentation. For ESL use, a teacher typically has to estimate the rough CEFR equivalent of a given grade-level output themselves.
Diffit levels content by grade band and was built for general K-12 differentiation, while tefl.ai’s AI Materials Adaptor is built specifically around CEFR levels for language learners. Teachers working inside a CEFR-based scheme of work often find the tefl.ai tool needs less manual conversion.
No. Diffit generates reading materials, vocabulary, and questions from a source text, but it doesn’t sequence those pieces into a timed lesson with a warm-up and closing task. That structuring step still depends on the teacher or a separate lesson-planning tool.
The past simple is the English verb form used for an action or state completed at a finished time in the past. Most verbs form it by adding -ed, while a closed set of common verbs change their form instead. It is the ordinary narrative form of English and the first past form most learners meet.
The present perfect is the English verb form made with have or has plus a past participle, used to link a past action to the present moment. It covers experience up to now, a situation that began in the past and still continues, and a past event whose result matters now. It never takes a finished time reference.
