Qwen3 VL 30B A3B Instruct
by Alibaba (Qwen) · qwen/qwen3-vl-30b-a3b-instruct · #69 cheapest of 325 paid models
Prices updated Jul 22, 2026, 2:26 AM UTC · refreshed hourly
Qwen3-VL-30B-A3B-Instruct is a multimodal model that unifies strong text generation with visual understanding for images and videos. Its Instruct variant optimizes instruction-following for general multimodal tasks. It excels in perception...
Compare vs:Claude Sonnet 5Claude Opus 4.6Claude Haiku 4.5GPT-5.6 Sol
Pricing
Input
$0.13
per 1M tokens
Output
$0.52
per 1M tokens
Blended (3:1)
$0.228
3 input : 1 output
Cache read
—
per 1M cached tokens
Cache write
—
per 1M tokens
Cache & batch economics
Effective prices for Qwen3 VL 30B A3B Instruct (no cache-read rate published — hits billed as normal input).
| Cache hit rate | Effective blended $/1M | Example request* | vs no cache |
|---|---|---|---|
| 0% | $0.228 | $0.0138 | — |
| 50% | $0.228 | $0.0138 | −0% |
| 90% | $0.228 | $0.0138 | −0% |
*Example: 100K sticky context + 2K new input + 1K output. Batch discount is an approximate 50% off list rates (available for OpenAI / Anthropic / Google — not flagged for this creator). Confirm on the vendor’s pricing page.
What a request costs
| Workload | Input tokens | Output tokens | Cost / request | Cost / 1K requests |
|---|---|---|---|---|
| Short chat message | 500 | 300 | $0.000221 | $0.221 |
| Document summary | 8,000 | 1,000 | $0.00156 | $1.56 |
| Codebase question (RAG) | 30,000 | 2,000 | $0.00494 | $4.94 |
| Long-context analysis | 150,000 | 5,000 | $0.0221 | $22.10 |
Performance
Intelligence Index
10
AA composite quality
Coding Index
—
Math Index
72.3
Agentic Index
—
Output speed
126 tok/s
median
Time to first token
0.95s
median
Time to first answer
0.95s
after reasoning tokens
Artificial Analysis benchmarks
| Benchmark | Score |
|---|---|
| MMLU-Pro | 76.4% |
| GPQA Diamond | 69.5% |
| Humanity's Last Exam | 6.4% |
| LiveCodeBench | 47.6% |
| SciCode | 30.8% |
| AIME 2025 | 72.3% |
| IFBench | 33.1% |
| AA-LCR (Long Context Reasoning) | 23.7% |
| Terminal-Bench Hard | 6.1% |
| τ²-Bench (Telecom) | 19.0% |
Specs
Context window
262K
tokens
Max output
33K
tokens
Input modalities
text, image
Tokenizer
Qwen3