Can Qwen3 32B run locally?

Yes, the weights are open and available on Hugging Face. It runs on Ollama, LM Studio, llama.cpp, MLX-LM and KTransformers, though the 32.8B size means you will need significant VRAM or a quantised version.

What does Qwen3 32B cost via API, and what is the context window?

The API price is $0.08 per million input tokens and $0.28 per million output tokens. The native context is 32,768 tokens, extendable to 131,072 tokens using YaRN-based scaling.

How is Qwen3 32B different from QwQ-32B?

Qwen3 32B replaces the always-on reasoning of QwQ with a switchable design: thinking mode for hard problems, non-thinking mode for fast general chat. Qwen says reasoning performance in thinking mode surpasses QwQ while non-thinking mode outperforms Qwen2.5 instruct.

Qwen3 32B

1533°

#llm #qwen #reasoning #coding #agents #multilingual

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 | 131,072 token context | $0.08/M input, $0.28/M output. Qwen3-32B is a dense 32.8B parameter causal language model from the Qwen3 series, optimized for both complex reasoning and efficient dialogue. It supports seamless switching between a "thinking" mode for tasks like math, coding, and logical inference, and a "non-thinking" mode for faster, general-purpose conversation. The model demonstrates strong performance in instruction-following, agent tool use, creative writing, and multilingual tasks across 100+ languages and dialects. It natively handles 32K token contexts and can extend to 131K tokens using YaRN-based scaling.

Guides for this model

Ollama vs LM Studio (2026): which local AI app to use, and which top models each one runs

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 32B is a 32.8-billion-parameter open-weights model from Qwen that runs in two modes: a deliberate thinking mode for math, coding and logic, and a faster non-thinking mode for everyday conversation. The ability to switch between those two gears in a single model, combined with genuine agent capability and support for over 100 languages, makes it unusually versatile for its size. It suits developers and researchers who want a capable local model without paying frontier API prices.

Known for

Dual thinking/non-thinking modes switchable within a single model
Strong agent and tool-use performance among open-weight models
100+ language support with solid multilingual instruction-following
Runs locally via Ollama, LM Studio, llama.cpp and vLLM
Very low API cost at $0.08/M input and $0.28/M output tokens

Watch out for

Native context is 32K; the 131K figure requires YaRN scaling, which can degrade quality at long range.
At 32.8B parameters, local use demands substantial VRAM — plan for a high-end GPU or quantised weights.
Thinking mode adds latency and token cost; for simple tasks, remember to disable it.

Best for

Prompts to try

Debug a concurrency bug

I have a Python async function that occasionally raises 'RuntimeError: dictionary changed size during iteration' under load. The function iterates over a shared dict while other coroutines may update it. Identify the root cause, explain why it surfaces intermittently, and rewrite the function so it is safe under concurrent access without using a global lock.

Plan a multi-step agent

You have access to three tools: web_search(query), read_file(path), and write_file(path, content). A user wants a weekly competitive-intelligence report on a named rival company saved to report.md. Write a step-by-step agent plan that decides when to call each tool, handles the case where a search returns no useful results, and keeps the final file under 800 words.

Solve a logic puzzle

Five houses in a row are each painted a different colour. Each house is occupied by a person of a different nationality, who drinks a different beverage, smokes a different brand and owns a different pet. The Brit lives in the red house. The Swede keeps dogs. The Dane drinks tea. The green house is immediately to the left of the white house. The green house owner drinks coffee. Who owns the fish? Show your reasoning step by step.

Compare with

Qwen3.8 Max (0902)
API-only model
1,000,000 token context · $2.00/M input, $6.00/M output
Qwen3.7 Plus
API-only model
1,000,000 token context · $0.32/M input, $1.28/M output
Qwen3.6 Plus
API-only model
1,000,000 token context · $0.33/M input, $1.95/M output
Qwen3.6 Max Preview
API-only model
262,144 token context · $1.03/M input, $6.16/M output

More text models from qwen

Questions people ask

Can Qwen3 32B run locally?: Yes, the weights are open and available on Hugging Face. It runs on Ollama, LM Studio, llama.cpp, MLX-LM and KTransformers, though the 32.8B size means you will need significant VRAM or a quantised version.
What does Qwen3 32B cost via API, and what is the context window?: The API price is $0.08 per million input tokens and $0.28 per million output tokens. The native context is 32,768 tokens, extendable to 131,072 tokens using YaRN-based scaling.
How is Qwen3 32B different from QwQ-32B?: Qwen3 32B replaces the always-on reasoning of QwQ with a switchable design: thinking mode for hard problems, non-thinking mode for fast general chat. Qwen says reasoning performance in thinking mode surpasses QwQ while non-thinking mode outperforms Qwen2.5 instruct.

Captain's verdict
A rare open-weight model that earns its keep in both deliberate reasoning and quick chat, at a price that makes the API almost an afterthought.

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

Trinity Large Thinking

TEXT / LLM

AI Models for Image, Video & Text Generation

Qwen3 32B 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 reasoning coding agents multilingual local

Browse by Model Family