← All models
Flagship — best answers, highest costOpen Weights
Qwen2.5 VL 72B Instruct Pricing
Qwen · qwen-qwen2-5-vl-72b-instruct
Input / 1M tokens
$0.800
Output / 1M tokens
$1.00
Context window
128k tokens
≈ 171 pages of text
Max output
128k tokens
What it costs in practice
Typical request (1,200 in + 400 out tokens)$0.0014
1,000 requests / month$1.36/mo10,000 requests / month$13.60/mo100,000 requests / month$136.00/moEstimate your own workloadOpens the calculator with Qwen2.5 VL 72B Instruct preloaded — adjust volume and token counts there.
Price history
List price per 1M tokens since we started tracking in Jun 2026. Each step marks a price change — hover a point for the exact date and rate.
What it excels at
Performs well on document, chart, and GUI understanding tasks with 128k output context. Competitive results on multilingual and code-related vision-language benchmarks.
The business tradeoff
72B open-weights model requires high GPU memory for local inference. Real-world latency and throughput depend heavily on the chosen hosting stack.
Cheaper from Qwen
Head-to-head
Vendor list rates, as of Jul 15, 2026 · source: openrouter · per-request examples assume 1,200 input + 400 output tokens.