Alibaba Qwen · AI model API pricing
How much does Qwen-Flash cost?
The cheapest input tokens of any model in this table — half of Gemini 2.5 Flash-Lite's rate.
Rate card
| Input tokens | $0.05 / 1M |
| Cached input | $0.01 / 1M |
| Output tokens | $0.40 / 1M |
| Context window | — |
What that means per month
| Workload | Tokens / month | Cost / month |
|---|---|---|
| Prototype | 1M in · 0.3M out | $0.17/mo |
| Small app | 10M in · 3M out | $1.70/mo |
| Production app | 50M in · 15M out | $8.50/mo |
| High volume | 200M in · 60M out | $34/mo |
| Heavy platform | 1B in · 300M out | $170/mo |
Standard on-demand rates from Alibaba Qwen's official Model Studio pricing, last checked 2026-07-27. Rates shown are for prompts ≤256k tokens; from 256k to 1M, input is $0.25 and output $2 per 1M. Implicit context cache is automatic and cannot be disabled: cached input reads bill at 20% of the input rate and cache creation bills at the standard rate (no write premium). Opting into the explicit cache instead reads at 10% but bills creation at 125%. Monthly figures exclude caching and batch discounts. Confirm live pricing before committing.
Where Qwen-Flash sits on price
At the reference month of 10M input and 3M output tokens, Qwen-Flash costs $1.70, which makes it the cheapest of the 30 models tracked here. Nothing else on the board runs this workload for less; the next cheapest is Gemini 2.5 Flash-Lite at $2.20. Among budget-tier models it is the cheapest of 9.
Cache economics
Cached input on Qwen-Flash bills at $0.01 per 1M — 20% of its own input rate, against a 10% median across the models here that publish a cached-read price. 25 models discount cached reads more steeply than this. On the reference month, serving 80% of that input from cache takes the input side of the bill from $0.50 to $0.18, leaving output untouched at $1.20 — which is why a high reuse rate changes the ranking above rather than just shaving the total.
Cache reads only. Writes are billed separately by some providers and are excluded here — see the rate note above for Alibaba Qwen's terms. Try it at your own hit rate on the AI cost calculator.