Qwen3 VL 30B A3B Thinking
Summary
Qwen3 VL 30B A3B Thinking is an open-weight language model from Alibaba (Qwen), released on 6 Oct 2025. It accepts text and images and generates text, with a 262K-token context window and up to 33K output tokens per response. It is scheduled for retirement on 15 Nov 2026.
Qwen3-VL-30B-A3B-Thinking is a multimodal model that unifies strong text generation with visual understanding for images and videos. Its Thinking variant enhances reasoning in STEM, math, and complex tasks. It excels...
Last verified 18 Nov 2026. Every figure on this page links to its source.
Specifications
| Developer | Alibaba (Qwen) |
|---|---|
| Release date | 6 Oct 2025 [OpenRouter] |
| Status | Generally available |
| Weights | Open weights [Hugging Face] |
| License | apache-2.0 [Hugging Face] |
| Parameters | 31.1B [Hugging Face] |
| Context window | 262K (262,144 tokens) [OpenRouter] |
| Max output | 33K tokens [OpenRouter] |
| Knowledge cutoff | 31 Mar 2025 [OpenRouter] |
| Input | Text and images [OpenRouter] |
| Output | Text [OpenRouter] |
| Reasoning mode | Yes [OpenRouter] |
| Tool calling | Yes [OpenRouter] |
| Structured output | Yes [OpenRouter] |
| Retirement date | 15 Nov 2026 [OpenRouter] |
| Hugging Face | Qwen/Qwen3-VL-30B-A3B-Thinking [Hugging Face] |
Qwen3 VL 30B A3B Thinking API pricing by provider
USD per million tokens. Sorted by input price.
We do not currently track a paid API listing for this model.
Running Qwen3 VL 30B A3B Thinking locally
Estimated memory, from the parameter count and the architecture in config.json on Hugging Face. Includes about 0.8 GB of runtime overhead.
| Quantization | Weights | Total, 8K context | Total, 32K context |
|---|---|---|---|
| Q4_K_M (4-bit) | 19 GB | 20 GB | 23 GB |
| Q8_0 (8-bit) | 33 GB | 35 GB | 37 GB |
| FP16 / BF16 (16-bit) | 62 GB | 64 GB | 66 GB |
At 4-bit with an 8K conversation, Qwen3 VL 30B A3B Thinking needs about 20 GB, so it fits on a graphics card with 24 GB or more, such as the NVIDIA GeForce RTX 3090 24 GB. Check your own hardware, or see Best GPU for local LLMs on our sister site AIPCs.
Frequently asked questions
What is the context window of Qwen3 VL 30B A3B Thinking?
Qwen3 VL 30B A3B Thinking has a context window of 262,144 tokens (262K) and can generate up to 32,768 tokens in a single response, according to OpenRouter.
Is Qwen3 VL 30B A3B Thinking open source?
Qwen3 VL 30B A3B Thinking is an open-weight model: its weights are published on Hugging Face as Qwen/Qwen3-VL-30B-A3B-Thinking under the apache-2.0 license. Check the license terms before commercial use.
When was Qwen3 VL 30B A3B Thinking released?
Alibaba (Qwen) released Qwen3 VL 30B A3B Thinking on 6 Oct 2025, according to OpenRouter.
How much VRAM does Qwen3 VL 30B A3B Thinking need?
About 20 GB at 4-bit (Q4_K_M) with an 8K-token conversation, and about 35 GB at 8-bit. The 4-bit model fits on a graphics card with 24 GB or more, such as the NVIDIA GeForce RTX 3090 24 GB. These are estimates from the model's published parameter count and architecture.
Similar models from other labs
Comparable on output type, weights, release date, price and context window. Prices are the lowest listed, USD per million tokens.
| Model | Lab | Released | Context | Input $/1M | Output $/1M | Weights |
|---|---|---|---|---|---|---|
| Kimi-K2-Instruct-0905 | Moonshot AI | 5 Sept 2025 | 262K | $0.60 | $2.50 | Open |
| Kimi-K2-Thinking | Moonshot AI | 6 Nov 2025 | 262K | $0.60 | $2.50 | Open |
| GLM-4.6 | Zhipu AI (Z.ai) | 30 Sept 2025 | 205K | $0.50 | $2 | Open |
| Command A Reasoning | Cohere | 21 Aug 2025 | 256K | $2.50 | $10 | Open |
| Ministral 14B | Mistral AI | 2 Dec 2025 | 262K | $0.20 | $0.20 | Open |
| Ministral 3 14B | Mistral AI | 2 Dec 2025 | 262K | $0.20 | $0.20 | Open |
Change history
- No longer listed on openrouter
- Retirement date changed from 9 Oct 2026 to 15 Nov 2026
- openrouter price changed: input $0.20 → $0.29, output $2.40 → $1 per 1M tokens