Qwen3.5 397B A17B API Pricing
Alibaba · Mid-tier · Qwen · released 2026-03. Prices in USD per 1 million tokens. Data snapshot: 2026-09-28.
The price we publish for Qwen3.5 397B A17B has changed 5 times since 2026-08-15, most recently on 2026-09-08 when the reference host changed (+49.0%); it is now the 44th cheapest of 108 Mid-tier models. Buying from Alibaba rather than at the rate quoted here lowers the token price by about 33%. It is rated a stable release but does not meet all four criteria of the FinOps Friendly badge.
- Input: $0.55 per 1M tokens
- Output: $3.50 per 1M tokens
- Batch API (input/output): not published
- Prompt-cache read: $0.23
- Context window: 262K
- Arena ELO: 1442 (as of 2026-09-25)
Qwen3.5 397B A17B market position
On blended price ($2.61 per 1M tokens at 30% input, 70% output) it is cheaper than 60% of Mid-tier models: 44th cheapest of 108. Across every category it is cheaper than 34% of priced text models in the catalogue: 186th cheapest of 281. Its Arena ELO of 1442 ranks 34th of the 68 models that carry one, above 51% of them.
Qwen3.5 397B A17B licence, size and capabilities
Qwen3.5 397B A17B is open weights, which is not the same as open source: the weights are published, but under Qwen, a licence that restricts some uses. Read it before self-hosting commercially. Its published size is 397B parameters. Listed capabilities: Text, Vision, Reasoning, Code, Agents. Filling its 262K context window once costs $0.14 in input tokens at list price.
Monthly cost at 100K requests, by business use case
From the after-discounts figure (prompt caching on the cached input share, batch API for async workloads, only where the provider publishes those rates), which assumes you implement both, to the list price. One figure means no discount is published.
| Use case | Tokens (in/out) | Monthly (after discounts – list) |
|---|---|---|
| Support Ticket | 1,500 / 500 | $228 – $258 |
| Knowledge Q&A | 2,000 / 800 | $345 – $390 |
| Meeting Summary | 10,000 / 1,200 | $938 – $970 |
| Marketing Content | 2,500 / 1,800 | $751 – $768 |
| Coding Task | 3,000 / 2,000 | $816 – $865 |
| Invoice Processing | 1,500 / 600 | $278 – $293 |
| Call Summary | 2,000 / 700 | $349 – $355 |
| Agent Workflow | 6,000 / 3,000 | $1.2K – $1.4K |
Qwen3.5 397B A17B pricing by host
A host is a company that runs the model and sells it by the token — the lab that made it, a cloud platform like Amazon Bedrock, Azure or Vertex, or an independent provider. One model, several sellers, and often several prices. This model is sold by 10 different hosts. The price quoted above is the one we publish so that models stay comparable; it is not necessarily the cheapest way to buy this model. List prices as published on 2026-09-28 via OpenRouter — negotiated rates, committed-use discounts and provisioned capacity are not shown.
Cheapest host: Alibaba at $0.39 input / $2.34 output per 1M tokens, 33% below the price we publish on a blended basis (30% input, 70% output). The dearest, Venice, charges 1.9x the cheapest.
| Host | Input $/1M | Output $/1M |
|---|---|---|
| Alibaba | $0.39 | $2.34 |
| DeepInfra — fp8 | $0.45 | $3.00 |
| DigitalOcean (the price shown above) | $0.55 | $3.50 |
| Phala (the price shown above) | $0.55 | $3.50 |
| AtlasCloud (the price shown above) — fp8 | $0.55 | $3.50 |
| Parasail — fp8 | $0.50 | $3.60 |
| StreamLake | $0.60 | $3.60 |
| GMICloud — fp8 | $0.60 | $3.60 |
| Novita | $0.60 | $3.60 |
| Venice | $0.75 | $4.50 |
Qwen3.5 397B A17B price history
Qwen3.5 397B A17B has been in our daily price archive since tracking began on 2026-08-15, when it listed at $0.39 input / $2.34 output per 1M tokens. The price we publish has changed 5 times since then, 3 of them a change of reference host rather than a repricing. The latest change, on 2026-09-08, was +49.0% on the blended price (30% input, 70% output), to $0.55 input / $3.50 output (reference host changed, not a repricing). Net of all changes, the blended price is +49.0% against 2026-08-15.
| Date | Old price (in / out) | New price (in / out) | Blended change |
|---|---|---|---|
| 2026-09-08 | $0.39 / $2.34 | $0.55 / $3.50 | +49.0% (reference host changed, not a repricing) |
| 2026-09-07 | $0.55 / $3.50 | $0.39 / $2.34 | -32.9% (reference host changed, not a repricing) |
| 2026-09-03 | $0.39 / $2.34 | $0.55 / $3.50 | +49.0% (reference host changed, not a repricing) |
| 2026-08-26 | $0.50 / $3.60 | $0.39 / $2.34 | -34.3% |
| 2026-08-24 | $0.39 / $2.34 | $0.50 / $3.60 | +52.1% |
Alternatives to Qwen3.5 397B A17B
- Next cheaper Mid-tier model: Kimi K2.6 (Moonshot) at $2.58 blended, 1% less.
- Next more expensive: Gemini 3.6 Flash (Google) at $2.85 blended, 9% more.
The nearest Mid-tier models on Arena ELO, each with its blended price and score difference against Qwen3.5 397B A17B.
- Qwen3.8 27B (Alibaba) — $0.42 in / $3.00 out per 1M, Arena ELO 1438: 15% cheaper, 4 ELO points lower
- GPT-5.1 (OpenAI) — $1.25 in / $10.00 out per 1M, Arena ELO 1438: 182% more expensive, 4 ELO points lower
- GPT-5.2 (OpenAI) — $1.75 in / $14.00 out per 1M, Arena ELO 1436: 295% more expensive, 6 ELO points lower
- GPT-5 (OpenAI) — $1.25 in / $10.00 out per 1M, Arena ELO 1435: 182% more expensive, 7 ELO points lower
- Cheapest within 15 Arena points: DeepSeek V4 Flash 0423 (DeepSeek), 94% cheaper, 6 ELO points lower, out of 15 models in that band.
- Highest Arena ELO within 20% of this price: Gemini 3.7 Flash (Google), 9% more expensive, 46 ELO points higher, out of 8 models in that price band.
Common questions about Qwen3.5 397B A17B pricing
- How much does Qwen3.5 397B A17B cost?
- Qwen3.5 397B A17B costs $0.55 per million input tokens and $3.50 per million output tokens on the API.
- Does Qwen3.5 397B A17B offer batch or cached pricing?
- Prompt-cache reads cost $0.23/M input tokens. No batch pricing is published.
- What would Qwen3.5 397B A17B cost per month?
- For a support-ticket workload (1,500 input + 500 output tokens per request) at 100K requests/month, Qwen3.5 397B A17B costs about $258 at list prices.
- Has the price of Qwen3.5 397B A17B changed?
- Qwen3.5 397B A17B has been in our daily price archive since tracking began on 2026-08-15, when it listed at $0.39 input / $2.34 output per 1M tokens. The price we publish has changed 5 times since then, 3 of them a change of reference host rather than a repricing. The latest change, on 2026-09-08, was +49.0% on the blended price (30% input, 70% output), to $0.55 input / $3.50 output (reference host changed, not a repricing). Net of all changes, the blended price is +49.0% against 2026-08-15.
These are list prices: real cost also depends on caching, batching, negotiated discounts, marketplace agreements, and existing contracts. Get a FinOps review from OptimNow or compare all models.
Build the business case for Qwen3.5 397B A17B in the AI ROI Calculator: add your business value on top of these costs to get ROI, break-even and payback.
All models · Compare side by side · Cloud compute pricing · Price barometer · FinOps guides · API and MCP server