Qwen3.5-27B API pricing
Alibaba · Mid-range price tier · Apache 2.0 · released 2026-03. Prices in USD per 1 million tokens. Data snapshot: 2026-11-12.
A support ticket workload of 100K requests a month costs $169.00 on Qwen3.5-27B; the next cheaper model in the Mid-range price tier for this workload, Devstral 2 2512, does it for $160.00. The price we quote for Qwen3.5-27B has changed 3 times since 2026-08-15, most recently on 2026-10-27 when the quoted host changed (+19.1%); it is now the 23rd cheapest of 113 models in the Mid-range price tier. Buying from Alibaba rather than at the price we quote lowers the token price by about 39%. It is rated a stable release but does not meet all four criteria of the FinOps Friendly badge.
- Input: $0.26 per 1M tokens
- Output: $2.60 per 1M tokens
- Batch API (in/out): Not published
- Prompt-cache read: Not published
- Context window: 262K tokens
- Arena ELO: 1409 as of 2026-11-08
Market position
On blended price ($1.898 per 1M tokens at 30% input, 70% output) it is cheaper than 80% of models in the Mid-range price tier: 23rd cheapest of 113. Across every price tier it is cheaper than 40% of priced text models in the catalogue: 183rd cheapest of 303. Its Arena ELO of 1409 ranks 92nd of the 160 models that carry one, above 43% of them. With an Arena score of 1409, Qwen3.5-27B sits below OptimNow's lowest working quality threshold, Standard (Arena 1420+), so the quality-bar method would not shortlist it for any of the 8 use cases the table prices. It is 11 points under the Standard bar (Arena 1420+), within the benchmark's 20-point noise. Each threshold is a distance below the best ranked model, which scores 1530 on Arena scores as of 8 Nov 2026.
Licence, size and capabilities
Qwen3.5-27B is open source: its weights are published under Apache 2.0, an OSI-approved licence with no field-of-use limit, so it can be self-hosted as well as bought by the token. Its published size is 27B parameters. Listed capabilities: Text, Vision, Reasoning, Agents. Filling its 262K context window once costs $0.0681 in input tokens at list price.
Pricing by host
A host is a company that runs the model and sells it by the token: the model maker, a cloud platform like Amazon Bedrock, Azure or Vertex, or an independent host. One model, several hosts, and often several prices. This model is sold by 6 different hosts. The price we quote elsewhere on this page is marked. We quote one host's rate so that models stay comparable; it is not a recommendation to buy there.
Cheapest host: Alibaba at $0.195 input / $1.56 output per 1M tokens, 39% below the price we quote on the blended price (30% input, 70% output). A support ticket workload at 100K requests a month costs $107.25 there against $169.00 at the price we quote, a difference of $61.75 a month. The dearest, DeepInfra, charges 1.6x the cheapest.
| Host | Input $/1M | Output $/1M | Precision |
|---|---|---|---|
| Alibaba (model maker) | $0.195 | $1.56 | not declared |
| Chutes | $0.20 | $1.75 | fp8 |
| SiliconFlow | $0.25 | $2.00 | fp8 |
| AtlasCloud | $0.27 | $2.16 | fp8 |
| Novita | $0.30 | $2.40 | bf16 |
| DeepInfra (price we quote) | $0.26 | $2.60 | fp8 |
List prices as published on 2026-11-12, via OpenRouter. Negotiated rates, committed-use discounts and provisioned capacity are not shown. These hosts do not all serve the model at the same numeric precision, so the cheapest is not a like-for-like substitute for the dearest.
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 Alibaba 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 | $169 |
| Knowledge Q&A | 2,000 / 800 | $260 |
| Meeting Summary | 10,000 / 1,200 | $572 |
| Marketing Content | 2,500 / 1,800 | $533 |
| Coding Task | 3,000 / 2,000 | $598 |
| Invoice Processing | 1,500 / 600 | $195 |
| Call Summary | 2,000 / 700 | $234 |
| Agent Workflow | 6,000 / 3,000 | $936 |
Price history
Qwen3.5-27B has been in our daily price archive since tracking began on 2026-08-15, when it listed at $0.195 input / $1.56 output per 1M tokens. The price we quote has changed 3 times since then, every time through a change of quoted host rather than a repricing. The latest change, on 2026-10-27, was +19.1% on the blended price (30% input, 70% output), to $0.26 input / $2.60 output (quoted host changed, not a repricing). Net of all changes, the blended price is +65.0% against 2026-08-15.
| Date | Old price (in / out) | New price (in / out) | Blended change |
|---|---|---|---|
| 2026-10-27 | $0.27 / $2.16 | $0.26 / $2.60 | +19.1% (quoted host changed, not a repricing) |
| 2026-10-19 | $0.26 / $2.60 | $0.27 / $2.16 | -16.1% (quoted host changed, not a repricing) |
| 2026-10-08 | $0.195 / $1.56 | $0.26 / $2.60 | +65.0% (quoted host changed, not a repricing) |
Alternatives to Qwen3.5-27B
1 of the 4 comparable models below costs less than Qwen3.5-27B: Mistral Large 4 undercuts it by 11%, and scores 1 Arena point higher.
- Next cheaper model in the Mid-range price tier: Nemotron 3 Ultra (Nvidia) at $1.86 blended, 2% less.
- Next more expensive: Seed 2.1 Turbo (ByteDance) at $1.90 blended, under 1% more.
The nearest models in the Mid-range price tier on Arena ELO, each with its blended price and score difference against Qwen3.5-27B.
- Mistral Large 4 (Mistral): $0.70 in / $2.10 out per 1M, Arena ELO 1410 (11% cheaper, 1 ELO point higher)
- o3 (OpenAI): $2.00 in / $8.00 out per 1M, Arena ELO 1410 (227% more expensive, 1 ELO point higher)
- GPT-5.4 Mini (OpenAI): $0.75 in / $4.50 out per 1M, Arena ELO 1412 (78% more expensive, 3 ELO points higher)
- GPT-5.2 (OpenAI): $1.75 in / $14.00 out per 1M, Arena ELO 1412 (444% more expensive, 3 ELO points higher)
- Cheapest within 15 Arena points: Qwen3.5-Flash (Alibaba), 89% cheaper, 12 ELO points lower, out of 29 models in that band.
- Highest Arena ELO within 20% of this price: DeepSeek V4 Pro 0423 (DeepSeek), 14% cheaper, 42 ELO points higher, out of 21 models in that price band.
Other models from Alibaba
The 8 nearest to Qwen3.5-27B in price, of 39 other Alibaba models in the catalogue. Prices per 1M tokens, input then output.
- Qwen3.8 27B: $0.15 / $1.875
- Qwen3 VL 235B A22B Instruct: $0.21 / $1.90
- Qwen3.6 Plus: $0.325 / $1.95
- Qwen3.5-122B-A10B: $0.26 / $2.08
- Qwen3.6 27B: $0.32 / $3.20
- Qwen3 235B A22B Thinking 2507: $0.45 / $3.50
- Qwen3.5 397B A17B: $0.60 / $3.60
- Qwen3.7 Max: $1.475 / $4.425
All 40 Alibaba models and prices
Common questions about Qwen3.5-27B pricing
- How much does Qwen3.5-27B cost?
- Qwen3.5-27B costs $0.26 per 1M input tokens and $2.60 per 1M output tokens at DeepInfra, the host whose rate we quote. DeepInfra serves a quantized fp8 build, which is not the same product as the full-precision model. The model maker, Alibaba, lists it at $0.195 per 1M input tokens and $1.56 per 1M output tokens. Prices as listed on 2026-11-12.
- Which host is cheapest for Qwen3.5-27B?
- Cheapest host: Alibaba at $0.195 input / $1.56 output per 1M tokens, 39% below the price we quote on the blended price (30% input, 70% output). A support ticket workload at 100K requests a month costs $107.25 there against $169.00 at the price we quote, a difference of $61.75 a month. The dearest, DeepInfra, charges 1.6x the cheapest. Host rates as listed on 2026-11-12.
- Does Qwen3.5-27B offer batch or cached pricing?
- No. Neither a batch rate nor a prompt-cache rate is published for Qwen3.5-27B as of 2026-11-12.
- What would Qwen3.5-27B cost per month?
- With the Support Ticket profile (1,500 input and 500 output tokens per request), 100,000 requests a month cost about $169 on Qwen3.5-27B at the list price quoted on 2026-11-12.
- Can Qwen3.5-27B be self-hosted?
- Yes. Qwen3.5-27B is open source: its weights are published under Apache 2.0, an OSI-approved licence with no field-of-use limit, so it can be self-hosted as well as bought by the token.
- Has the price of Qwen3.5-27B changed?
- Qwen3.5-27B has been in our daily price archive since tracking began on 2026-08-15, when it listed at $0.195 input / $1.56 output per 1M tokens. The price we quote has changed 3 times since then, every time through a change of quoted host rather than a repricing. The latest change, on 2026-10-27, was +19.1% on the blended price (30% input, 70% output), to $0.26 input / $2.60 output (quoted host changed, not a repricing). Net of all changes, the blended price is +65.0% against 2026-08-15.
Its weights are published, which changes what you may run yourself: the cost gap between open weights and proprietary APIs.
These are list prices. Your real cost also depends on caching and batching, negotiated discounts, marketplace agreements, and the contracts you already have. OptimNow audits exactly that.
Build the business case · Get a FinOps review
LLM pricing · Models by maker · Cloud compute pricing · Compare models · Price barometer · FinOps guides · API and MCP server