Qwen3.5 397B A17B API Pricing
Alibaba (Qwen) · context window 262K · prices updated 2026-09-15
Where Qwen3.5 397B A17B sits on price
At $3.50 per million output tokens, Qwen3.5 397B A17B is a mid tier model — 1.7× the median output price of $2.08, which puts it cheaper than 39% and pricier than 61% of the models we track. It sits 45th cheapest of the 53 Alibaba (Qwen) models we track. Output costs 6.4× more than input here, so anything that makes the model write less — tighter instructions, structured output, lower max_tokens — moves the bill more than trimming the prompt.
Cached input is 59% cheaper. On an input-heavy workload you need roughly a 42% cache-hit rate to take 25% off the input line — reachable for chatbots and agents that resend the same system prompt and history.
What real workloads cost on Qwen3.5 397B A17B
These three workloads are the ones teams actually run on a mid-priced model — a $3.50/1M model is not bought for the same job as one ten times the price.
| Workload | Per request | Per month |
|---|---|---|
| Support chatbot — 500 conversations/day (5K in / 1.4K out) Conversation history re-sent each turn — the classic prompt-caching win. | $0.0077 | $115 |
| Document pipeline — 10K docs/day (3K in / 500 out) Batch-friendly: no user waiting, so a ~50% batch discount usually applies. | $0.0034 | $1020 |
| Small app — 1K requests/day (500 in / 150 out) The baseline most side projects actually run at. | $0.0008 | $24.00 |
Model your exact traffic in the API cost calculator — it preloads Qwen3.5 397B A17B with caching and batch options.
Qwen3.5 397B A17B price history
Alibaba (Qwen) last raised Qwen3.5 397B A17B's output price on Sep 9, 2026 (+50% per output token), from $0.39/$2.34 to $0.55/$3.50 per 1M in/out. That is one of 9 repricings in the 44 days we have tracked it — worth knowing before you hard-code today's rate into a budget.
| Date | Input /1M | Output /1M | Output change |
|---|---|---|---|
| Aug 2, 2026 (tracking began) | $0.39 | $2.34 | — |
| Aug 9, 2026 | $0.5 | $3.60 | +54% |
| Aug 10, 2026 | $0.39 | $2.34 | -35% |
| Aug 11, 2026 | $0.5 | $3.60 | +54% |
| Aug 14, 2026 | $0.39 | $2.34 | -35% |
| Aug 24, 2026 | $0.5 | $3.60 | +54% |
| Aug 26, 2026 | $0.39 | $2.34 | -35% |
| Sep 3, 2026 | $0.55 | $3.50 | +50% |
| Sep 7, 2026 | $0.39 | $2.34 | -33% |
| Sep 9, 2026 | $0.55 | $3.50 | +50% |
Change-points from our daily price snapshots (tracking since Aug 2, 2026; intraday moves between snapshots are not captured).
Qwen3.5 397B A17B vs Kimi K2.7 Code
The closest-priced alternative from another vendor is Kimi K2.7 Code (Moonshot (Kimi)) — priced within a rounding error on output, with $0.16 more per million input tokens. When two models land this close on price, the decision is quality on your own workload, not the price sheet: run 50 real requests through both and compare.
See Kimi K2.7 Code pricing →Cheaper alternatives
More Alibaba (Qwen) models
Frequently asked questions
› How much does the Qwen3.5 397B A17B API cost?
Qwen3.5 397B A17B costs $0.55 per million input tokens and $3.50 per million output tokens, with cached input at $0.225 per million (59% cheaper). That works out to roughly 285,714 output tokens per dollar.
› What does the support chatbot workload cost on Qwen3.5 397B A17B?
Support chatbot — 500 conversations/day (5K in / 1.4K out) costs about $0.0077 per request and $115 per month on Qwen3.5 397B A17B. Conversation history re-sent each turn — the classic prompt-caching win.
› What does it cost to fill Qwen3.5 397B A17B's 262K context window?
Sending 262K of input in a single request costs $0.1442 at $0.55 per million tokens — before any output. With prompt caching that same fill drops to about $0.059 on repeat requests. This is why large context windows are cheap to advertise and expensive to actually use.
› Is Qwen3.5 397B A17B worth the price?
Qwen3.5 397B A17B sits in the middle of the market (45 of 53 by price within Alibaba (Qwen)). The honest test is a routing experiment: send the same 200 real requests to this model and to a tier below, and compare failure rate against the price gap — most teams find a majority of traffic never needed the pricier model.