Qwen3 VL 32B Instruct
Qwen · released Oct 23, 2025
Qwen3-VL-32B-Instruct is a large-scale multimodal vision-language model designed for high-precision understanding and reasoning across text, images, and video. With 32 billion parameters, it combines deep visual perception with advanced text...
Specification
- Context window
- 131K
- Max output
- 33K
- Knowledge cutoff
- Not stated
- Parameters
- 33.4B
- Licence
- apache-2.0
- Serving providers
- 1
- Moderated
- No
- Uptime
- 100.0%
Intelligence
11.0
38th percentile
Coding
—
Coding Index
Agentic
—
Agentic Index
Output speed
47 t/s
Median across providers
Latency
837ms
Time to first token
Cost per task
$0.02
Estimated
Benchmarks
Where the score comes from
The Intelligence Index is a composite. These are the underlying evaluations this model was actually measured on.
Evaluation scores
Percentage correct · higher is better
- MMLU-Pro79.1%
- AIME 202568.3%
- GPQA Diamond67.1%
- LiveCodeBench51.4%
- IFBench39.2%
- AA-LCR (long context)34.3%
- SciCode30.1%
- τ²-bench (Telecom)29.2%
- Terminal-Bench Hard8.3%
- Humanity's Last Exam6.8%
An evaluation missing from this list was not run for this model — it is not a zero.
View as table
| Evaluation | Score |
|---|---|
| MMLU-Pro | 79.1% |
| AIME 2025 | 68.3% |
| GPQA Diamond | 67.1% |
| LiveCodeBench | 51.4% |
| IFBench | 39.2% |
| AA-LCR (long context) | 34.3% |
| SciCode | 30.1% |
| τ²-bench (Telecom) | 29.2% |
| Terminal-Bench Hard | 8.3% |
| Humanity's Last Exam | 6.8% |
Against its peers
Intelligence Index · this model highlighted, nearest peers in grey
- GPT-4o (2024-11-20)OpenAI11.1
- GPT-4oOpenAI11.1
- Qwen3 VL 32B InstructQwen11.0
- GLM 4.6VZ.ai10.9
- Llama 4 ScoutMeta10.3
- 9.9
- R1 Distill Llama 70BDeepSeek9.8
- GPT-4.1 NanoOpenAI9.6
Peers are the models sitting closest on the Intelligence Index — the set you would realistically choose between.
View as table
| Model | Intelligence |
|---|---|
| GPT-4o (2024-11-20) | 11.1 |
| GPT-4o | 11.1 |
| Qwen3 VL 32B Instruct | 11.0 |
| GLM 4.6V | 10.9 |
| Llama 4 Scout | 10.3 |
| Qwen3 VL 30B A3B Instruct | 9.9 |
| R1 Distill Llama 70B | 9.8 |
| GPT-4.1 Nano | 9.6 |
Percentile among all indexed models
Pricing
What it costs to run
List prices per million tokens, plus what one representative task works out to.
List price
- Input / 1M tokens
- $0.104
- Output / 1M tokens
- $0.416
- Cached input / 1M
- Not offered
- Blended 3:1
- $0.182
One task, estimated
$0.02
- Input tokens
- 50,000
- Output tokens
- 25,000
- Profile
- Standard
Estimated from list pricing: 50K input tokens plus 80K output tokens for reasoning models (25K for non-reasoning).
Serving providers
Speed and latency figures are medians across these providers, so a widely-served model reports a blend rather than any single endpoint.