Alibaba (Qwen) · language model
Qwen3 32B: pricing, context and benchmarks
Qwen3 32B is a language model from Alibaba (Qwen) released on April 28, 2025. It costs $0.08 per million input tokens and $0.28 per million output tokens, with a 131.1K-token context window. Epoch AI ranks it #105 of 125 rated models.
| Performance (ECI) | 7/40 | |
|---|---|---|
| Price | 26/30 | |
| Context | 4/15 | |
| Features | 15/15 |
What you need to know about Qwen3 32B
Qwen3 32B was released by Alibaba (Qwen) (China) on April 28, 2025. It accepts text as input and produces text. Its context window reaches 131,072 tokens, larger than 29 % of the models in the registry, and it can write up to 16,384 tokens in one go.
It has been available for about 18 months, long enough for its prices and scores to have settled.
Qwen3 32B pricing
On OpenRouter, Qwen3 32B costs $0.08 per million input tokens and $0.28 per million output tokens. Counting three tokens read for every token written, its average price is $0.13 per million, more expensive than 14 % of paid models (5.9 times cheaper than the median of $0.76).
A concrete example: a task that sends 2,000 tokens (about 1,500 words) and receives a 500-token answer costs $0.0003. At 1,000 tasks a day, budget $9 a month. The price calculator redoes this with your own volumes.
Performance and benchmarks
The Epoch Capabilities Index (ECI), which aggregates published results across many benchmarks, gives Qwen3 32B a score of 138.4 (range 134.7 to 140.2). That places it #105 of 125 rated models.
Scores published by Epoch AI on the benchmarks the site tracks:
| Benchmark | What it measures | Score | Date |
|---|---|---|---|
| GPQA Diamond | PhD-level science questions | 65.7 % | Apr 29, 2025 |
| OTIS Mock AIME | maths competition problems | 66.9 % | Apr 29, 2025 |
| Aider Polyglot | coding exercises in several languages | 40 % | Apr 29, 2025 |
For at least equal performance, DeepSeek V4 Flash 0731, Qwen3.7 Flash, and gpt-oss-20b cost at least 30% less.
Features, privacy and local use
- It supports tool calling, so it can drive an AI agent, call APIs or MCP servers.
- It can answer in an enforced JSON format (structured outputs).
- It is a reasoning model: it "thinks" before answering, which helps on hard tasks but uses more output tokens.
- Its knowledge stops at March 2025.
Its weights are open and downloadable from Hugging Face (apache-2.0 licence): you can run it on your own machines without sending your data to a third party. With 32,762,123,264 billion parameters, it needs about 18,307,262,993 GB of memory at 4-bit quantisation. Check with the VRAM calculator. It was downloaded 2,880,776 times on Hugging Face last month.
Other Alibaba (Qwen) models
| Model | Released | Avg. price | ECI |
|---|---|---|---|
| Qwen3.8 Max Prime | Sep 23, 2026 | $6 | — |
| Qwen3.8 Omni Flash | Sep 21, 2026 | $0.23 | — |
| Qwen3.8 Max (0902) | Sep 3, 2026 | $3 | 155.1 |
| Qwen3.8 Flash | Aug 26, 2026 | $0.23 | — |
| Qwen3.8 27B | Aug 14, 2026 | $0.78 | 151.7 |
| Qwen3.8 2.4T A95B | Aug 12, 2026 | $3 | — |
| Qwen3.7 Flash | Jul 27, 2026 | $0.06 | 144.1 |
| Qwen3.7 Plus | Jun 3, 2026 | $0.56 | 147 |
Verdict
Qwen3 32B scores 52/100 on our grid, which only uses published figures.
Strengths
- Low price: cheaper than 86 % of paid models.
- Open weights: self-hostable, your data stays with you.
- Agent-ready: tool calling and structured JSON output.
Limitations
- Modest performance: behind 83 % of rated models.
- Text only: reads neither images nor files.
Avoid if…
- you want to host it on a single consumer computer (≈ 18,307,262,993 GB needed at 4-bit).
Alternatives
- Gemma 3 27B$0.15
- gpt-oss-20b$0.04
- gpt-oss-120b$0.26
- GPT-5 Nano$0.14
- Gemma 3 12B$0.08
- DeepSeek V3.1$0.43
Comparisons and tools
Learn more
Frequently asked questions
How much does Qwen3 32B cost?
$0.08 per million input tokens and $0.28 per million output tokens on OpenRouter (checked October 25, 2026). A task with 2,000 tokens read and 500 written costs $0.0003.
How big is Qwen3 32B's context window?
131,072 tokens, roughly 98,000 words, with a maximum of 16,384 output tokens.
Can Qwen3 32B run locally?
Yes, its weights are published under the apache-2.0 licence. Expect about 18,307,262,993 GB of memory at 4-bit.
Is Qwen3 32B suitable for an AI agent?
Technically yes: it supports tool calling and structured outputs. Then check its cost for your volume.
How good is Qwen3 32B?
Epoch AI ranks it #105 of 125 rated models (ECI 138.4). On GPQA Diamond, it scores 65.7%.
What are the alternatives to Qwen3 32B?
The closest models on price and performance are Gemma 3 27B, gpt-oss-20b, and gpt-oss-120b.
Related articles
- Mistral AI: models, pricing and agentsOctober 25, 2026
- LLM benchmarks: how to read model rankingsOctober 24, 2026
- Running a local LLM: hardware, software, and limitationsOctober 24, 2026
- Open-source or closed models for your AI agents?October 5, 2026
Neutrality. No affiliate links. No vendor pays to be reviewed, rated or ranked. Reviews are directed by François Chollet, neutral and factual, and every data point is dated and sourced. How we score