Can I run Qwen3.5 397B-A17B locally?

Yes, the weights are open and published on Hugging Face under the Qwen/Qwen3.5-397B-A17B repository, compatible with vLLM, SGLang and KTransformers. With 397B total parameters you will need substantial hardware; most people running it locally will use a quantised version.

What does it cost and how long is the context?

Via API the model is priced at $0.55 per million input tokens and $3.50 per million output tokens, with a native context of 262,144 tokens. The hosted Qwen3.5-Plus variant extends this to 1 million tokens by default.

How does this differ from Qwen3 or Qwen3-VL?

Qwen3.5 uses early-fusion multimodal training, meaning vision and language are unified from the start rather than bolted together, and the maker states it reaches cross-generational parity with Qwen3 while outperforming Qwen3-VL models on reasoning, coding and visual understanding benchmarks.

Qwen3.5 397B A17B

4310°

#llm #qwen #vision #video #reasoning #agents

Run this model

Offers / Ads

OpenRouter Terminal / API

Pay per token

Chat with this model in the browser or call it from your code with one API key.

Qwen | 262,144 token context | $0.50/M input, $3.60/M output. The Qwen3.5 series 397B-A17B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. It delivers state-of-the-art performance comparable to leading-edge models across a wide range of tasks, including language understanding, logical reasoning, code generation, agent-based tasks, image understanding, video understanding, and graphical user interface (GUI) interactions. With its robust code-generation and agent capabilities, the model exhibits strong generalization across diverse agent.

Guides for this model

Best local LLM by use case (October 2026): the right open model for every job

How much VRAM do I actually need for local AI?

What is a GGUF file? Quantized models explained (Q4, Q8, fp8, NF4)

All guides

Captain's notes

Written by Captain, the house editor

Notes from 2026-10-13

Qwen3.5 397B-A17B is Alibaba's flagship open-weights vision-language model: 397 billion total parameters, but only 17 billion active per token thanks to a sparse mixture-of-experts design paired with a linear attention mechanism. It handles text, images, video and GUI interactions in a single model, with native 262K context and support for 201 languages. It suits developers who want frontier-class multimodal capability without paying closed-API prices.

Known for

Sparse MoE keeps only 17B parameters active, cutting inference cost relative to model size.
Native vision, video and GUI understanding built in from early-fusion training.
Strong code generation and agent task performance across diverse environments.
201 languages and dialects supported, including regional and cultural nuance.
Open weights on Hugging Face, compatible with vLLM, SGLang and KTransformers.

Watch out for

397B total weights demand serious hardware to self-host; most users will hit the API instead.
Output tokens cost $3.50/M — not cheap at scale, especially for long agentic loops.
The hosted Qwen3.5-Plus version adds 1M context and built-in tools; the open weights cap at 262K natively.

Best for

Prompts to try

Debug a failing agent loop

I have a Python agent that calls a web-search tool, parses the result, then decides whether to call the tool again or return a final answer. It sometimes loops indefinitely. Here is the code: [paste your code]. Identify why the termination condition fails and rewrite the loop so it exits correctly after at most five iterations.

Analyse a screenshot for bugs

Attach a screenshot of your application's UI. Identify every visible error state, broken layout element or confusing label. For each issue, explain what is wrong and suggest a concrete fix a developer could implement in under an hour.

Multilingual contract clause

Translate the following limitation-of-liability clause into French, German, Japanese and Brazilian Portuguese. After each translation, add one sentence noting any legal concept in the source text that does not map cleanly to that jurisdiction's standard practice. Source clause: 'In no event shall either party be liable for indirect, incidental or consequential damages arising from use of this service.'

Compare with

Qwen3.7 Flash
API-only model
1,000,000 token context · $0.03/M input, $0.13/M output
Qwen3.8 Max (0902)
API-only model
1,000,000 token context · $2.00/M input, $6.00/M output
Qwen3.8 Omni Flash
API-only model
1,000,000 token context · $0.15/M input, $0.47/M output
Qwen3.5 Plus 2026-02-15
API-only model
1,000,000 token context · $0.26/M input, $1.56/M output

More text models from qwen

Questions people ask

Can I run Qwen3.5 397B-A17B locally?: Yes, the weights are open and published on Hugging Face under the Qwen/Qwen3.5-397B-A17B repository, compatible with vLLM, SGLang and KTransformers. With 397B total parameters you will need substantial hardware; most people running it locally will use a quantised version.
What does it cost and how long is the context?: Via API the model is priced at $0.55 per million input tokens and $3.50 per million output tokens, with a native context of 262,144 tokens. The hosted Qwen3.5-Plus variant extends this to 1 million tokens by default.
How does this differ from Qwen3 or Qwen3-VL?: Qwen3.5 uses early-fusion multimodal training, meaning vision and language are unified from the start rather than bolted together, and the maker states it reaches cross-generational parity with Qwen3 while outperforming Qwen3-VL models on reasoning, coding and visual understanding benchmarks.

Captain's verdict
A rare open-weights model that genuinely competes on vision, agents and code at once — if you can afford the output tokens or the hardware.

Model card on Hugging Face

Captain writes these notes from the model's own description and model card. Prices and context sizes are the provider's at the time of writing and can change.

Ranking badge

Updated Oct 22, 2026

Made this model? Show this badge on your page.

Paste it into your Hugging Face card or Civitai description.

Change colors / personalize / get code

Not this creator?

Get your own badge

How this score adds up

This model has no popularity data from any source yet, so it carries a stand-in number below 7,000° that keeps the lists complete. It never ranks above a model with data. The breakdown appears here as soon as a source reports it.

Glossary: the terms used here

Heat score: One number for how much a model is used and wanted right now, from 0° to 9,999°. Every list on this site that says hottest is sorted by it and nothing else.
Floor: The 7,000° every model with real data starts from. The scale runs from 0° to 9,999°; the part below the floor is reserved for stand-ins (models without data), so that a model with data always ranks above one without. The floor is the same for everyone and decides nothing; the degrees above it are the measurement.
Download-equivalents: Everything the score treats as a download, added up: downloads themselves, each torrent seeder as 1,000, each AI Horde image and each run on a generation site as one.
Log scale: Counting by tens, not by ones: going from 1,000 to 10,000 downloads adds the same degrees as going from 10,000 to 100,000. It keeps giants from crowding out everything else.
Likes: Thumbs up and likes from every source, worth half a download each on the log scale.
Adoption rate: How many downloads a day a model gets right now, from the difference between two readings of a source. A quarter of the weight of the download term.
Measured and estimated: Measured means two readings of a source exist and the difference is real. Estimated means only today's total is known and the past is spread over the model's life; the site says which one it is showing.
Seeder: Someone who keeps a copy of the weights online for others to download over a torrent, or a hosted mirror doing the same. Each one counts as 1,000 downloads.
AI Horde: A volunteer network that generates images for free on donated computers. The number of images it has made with a model is public, and each counts as one download.
Generation sites: Sites that host their own copies of a model and let people generate with it, such as PixAI, SeaArt and Yodayo. Their downloads count like Civitai's and each generation run as one download.
OpenRouter tokens: A text model served through an API, such as Claude or GPT, has no weights to download, so its usage is measured in tokens: the pieces of text it read and wrote through OpenRouter, a gateway many apps use, over the last 30 days. Every 10,000,000 tokens count as one download. Source: OpenRouter (openrouter.ai/rankings), CC BY 4.0.
Full model (checkpoint): The weights themselves, the file that generates on its own. It gets a bonus because add-ons depend on it.
LoRA and embedding: Small add-on files that steer someone else's weights toward a style, a character or a concept. They cannot run alone.
Family and weights: The base a model is built on, such as SD 1.5, SDXL, Flux or Wan. Every model inherits the release date of its family's weights.
Age: Degrees taken off for how long ago the family's weights came out: 150° a year, up to 750°. It keeps old downloads that had years to pile up from freezing the top of the list.
Family boost: A small, bounded nudge for LoRAs of a family this site is pushing forward right now, always listed in the breakdown.
Stand-in score: A stable number below the floor given to a model no source reports on yet, so lists stay complete. It never ranks above a model with data.
New! (grace period): A model less than 7 days old shows New! instead of a score and holds no place in the rankings. Its score is computed from day one, but a model one day old grows faster than anything, so without the wait the week's #1 would always be whatever came out yesterday rather than what held up for a week.
Trend: How a model's growth over a period compares with the typical model's growth over the same period. Faster than typical reads as a rise, slower as a fall.
Typical model: The middle of the catalog: half the models with a figure grew faster over the period, half slower.

Similar Models

AI Models for Image, Video & Text Generation

Qwen3.5 397B A17B is a text model for the Text / LLM family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Text / LLM models and check its heat score to see how it is trending.

More Text / LLM models llm qwen vision video reasoning agents coding multilingual

Browse by Model Family