See how ChatGPT, Perplexity and Google AI Overviews describe you today
Free AI Visibility AuditQwen3.5 Plus 2026-04-20 is a large-scale multimodal language model from Alibaba's Qwen family that takes text, image and video as input and returns text, with a context window that runs to a million tokens.
QwenVerified 25/09/2026Released 27/04/202601 / decision
Each figure sits next to the middle of the field, so it can be read as dear or cheap, wide or narrow, rather than floating on its own. Compared against the 231 models we track, not against an absolute standard.
Capability
Not measured
No published benchmark scores yet
Input / 1M tokens
$0.30
field median $0.42
Output / 1M tokens
$1.80
field median $1.81
Context
1,000K
field median 500K
02 / overview
Four questions, answered separately, because somebody arrives at one of them rather than at the top of the page.
The job it was built for, and the job it is not for.
Qwen3.5 Plus is a general-purpose multimodal workhorse: one model that reads documents, images and video clips and writes text back. It is built for the broad middle of production work, summarising, extracting, answering and drafting across mixed inputs, rather than for any single specialist task. It does not generate images or video, only interpret them.
Where it sits in its line, and when a sibling is the better pick.
This April 2026 build of Qwen3.5 Plus was first seen in the wild on 27 April 2026, sitting in the Plus tier of the Qwen line, the general-duty rung rather than the small fast one or the top reasoning one. Reach for it when a job mixes long documents with images or video and you want a single model to handle all of it. If your workload is short, text-only and high-volume, a smaller Qwen tier will do the same work for less.
The API, the apps it powers, and what its limits let you do.
You reach it through the Qwen API, and through Alibaba Cloud's model services where the Qwen line is served. The million-token context means an entire contract set, codebase or a long video transcript with its frames can go in on one call without chunking, and the large output ceiling means the reply can be a full report rather than a summary of one. Because image and video go in alongside text, workflows like reading a screen recording against a spec, or checking product photography against a brand guide, run in one pass.
What changes because this exists, or why it does not.
The combination that matters here is a very long context plus video input at a low per-token rate, which makes it practical to throw whole messy inputs at a model instead of building a pipeline to trim them first. Teams that were splitting documents, transcribing video separately and stitching results together can collapse that into a single call. That is an engineering simplification more than a capability leap.
Follows Qwen3.6 Plus. Superseded by Qwen3.6 27B. See the whole line.
03 / evidence
Split by category, so a strong number never hides a thin evidence base. Verified means we read it on the benchmark's own published results; a provider's claim about its own model is shown and labelled rather than dropped.
No published benchmark scores for this model yet.
We publish a score only where we can link the result to where it was published. Until a benchmark result for this model exists in a source we read, this section stays empty rather than being filled with a provider's marketing figure.
How we decide what counts as evidence04 / ledger
Every published row, grouped by category, each compared with the best published score on the same benchmark. 'Is 64% good' is a question nobody can answer; '26 points behind the leader' is one anybody can.
Nothing in the ledger yet.
Each row here carries a score, the benchmark it came from, what the leading model scored on the same test, and a link to the published result. Rows appear as results are published and read.
How we decide what counts as evidence05 / capability
Where this model is strong, and against how many peers. Ranks are against models with evidence in that category, not against everything we track: ranking against models nobody tested would rank who published, not who is better.
No category scores to shape yet.
A category score is the weighted mean of the benchmarks published for it. With no published rows there is nothing to average, and an empty chart drawn at zero would say something false.
How we decide what counts as evidence06 / cost
List API rates as last read from the provider, with the source on every row, plus every change we have recorded since we started tracking it.
| Charge | Price | Unit | Read on | Source |
|---|---|---|---|---|
| Input | $0.30 | per 1M tokens | 2026-09-24 | Check |
| Output | $1.80 | per 1M tokens | 2026-09-24 | Check |
It is priced at the cheap end of the multimodal field, with output costing several times input as is normal, and the whole thing sitting well below what frontier Western models charge for comparable context. For a long-context multimodal model, the cost profile makes bulk document and video processing viable as a background job rather than a premium feature.
| Workload | Input / month | Output / month | Cost |
|---|---|---|---|
| A small product team | 20M tokens | 5M tokens | $15.00 |
| A busy support assistant | 200M tokens | 40M tokens | $132.00 |
| A document pipeline | 1000M tokens | 100M tokens | $480.00 |
List API rates, no caching and no batch discount, which both providers offer and which change the answer a great deal. Treat these as the ceiling, not the bill.
07 / specs
As listed by the provider's own catalogue and re-read every few hours. Anything absent is absent there too.
| Context window | 1,000,000 tokens |
|---|---|
| Maximum output | 65,536 tokens |
| Modalities | text, image, video |
| Released | 27/04/2026 |
| Status | Current |
| Catalogue identifier | qwen/qwen3.5-plus-20260420 |
08 / lineage
What this model replaced, what replaced it, and what else its provider has in the field.
Also from Qwen
09 / line
Every model in this family in release order, so a page from eight months ago says in one glance that two newer ones exist.
Came before
← Qwen3.6 PlusCame after
Qwen3.6 27B →Ordered by release date and worked out from the naming, so a new member slots in as soon as its page exists. A retired model keeps its page and its place in the line.
10 / notes
What this model changes for a brand trying to be cited in AI answers, and every change we have logged since it launched.
What it changes for you
Qwen is the model layer behind a large share of AI answers in Chinese-language and Asia-Pacific markets, so if that is where your buyers are, your visibility there depends on what Qwen has retrieved and retained about you. The long context means it can hold a lot of your site at once, which rewards depth and consistency over thin pages. Check how you are described in Qwen answers separately from how you are described in ChatGPT or Gemini, the sources and the wording diverge.
Buyers meet Qwen models through the Qwen chat app and through Alibaba Cloud's API, which is the default route for teams building in China and across South East Asia. Qwen also shows up inside third-party tools and assistants that route to it for cost reasons, often without naming it in the interface. If your buyers are in those markets, answers they read may well have been written by this model.
Nothing published here yet. Changes appear within hours of a provider announcing them.
11 / questions
The things people ask about this model, answered from what is on this page rather than from anywhere else.
Models change what gets cited. We measure whether AI answers name your brand or your competitors across every assistant, and show you what to change.