Qwen3.5-35B-A3B API pricing
Alibaba · Budget 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 $71.00 on Qwen3.5-35B-A3B; the next cheaper model in the Budget price tier for this workload, Mi:dm K 3.0 Pro, does it for $70.00. The price we quote for Qwen3.5-35B-A3B has changed 12 times since 2026-08-15, most recently on 2026-10-21 when the quoted host changed (-0.4%); it is now the 112th cheapest of 160 models in the Budget price tier. Buying from Darkbloom rather than at the price we quote lowers the token price by about 27%, for a quantized fp4 build. It is rated a stable release but does not meet all four criteria of the FinOps Friendly badge.
- Input: $0.14 per 1M tokens
- Output: $1.00 per 1M tokens
- Batch API (in/out): Not published
- Prompt-cache read: $0.05 per 1M tokens
- Context window: 262K tokens
- Arena ELO: 1395 as of 2026-11-08
Market position
On blended price ($0.742 per 1M tokens at 30% input, 70% output) it is cheaper than 30% of models in the Budget price tier: 112th cheapest of 160. Across every price tier it is cheaper than 63% of priced text models in the catalogue: 112th cheapest of 303. Its Arena ELO of 1395 ranks 103rd of the 160 models that carry one, above 36% of them. With an Arena score of 1395, Qwen3.5-35B-A3B 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. 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
Qwen3.5-35B-A3B 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 35B parameters. Listed capabilities: Text, Vision, Reasoning, Agents. Filling its 262K context window once costs $0.0367 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 7 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: Darkbloom at $0.065 input / $0.75 output per 1M tokens, 27% below the price we quote on the blended price (30% input, 70% output). It serves a quantized fp4 build, which is not the same product as the full-precision model. A support ticket workload at 100K requests a month costs $47.25 there against $71.00 at the price we quote, a difference of $23.75 a month. The dearest, SiliconFlow, charges 2.4x the cheapest.
| Host | Input $/1M | Output $/1M | Precision |
|---|---|---|---|
| Darkbloom | $0.065 | $0.75 | fp4 |
| DeepInfra (price we quote) | $0.14 | $1.00 | fp8 |
| Parasail | $0.15 | $1.00 | fp8 |
| Venice | $0.15 | $1.00 | not declared |
| Alibaba (model maker) | $0.1625 | $1.30 | not declared |
| AtlasCloud | $0.225 | $1.80 | fp8 |
| SiliconFlow | $0.24 | $1.80 | 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 | $63 – $71 |
| Knowledge Q&A | 2,000 / 800 | $95 – $108 |
| Meeting Summary | 10,000 / 1,200 | $251 – $260 |
| Marketing Content | 2,500 / 1,800 | $211 – $215 |
| Coding Task | 3,000 / 2,000 | $229 – $242 |
| Invoice Processing | 1,500 / 600 | $77 – $81 |
| Call Summary | 2,000 / 700 | $96 – $98 |
| Agent Workflow | 6,000 / 3,000 | $346 – $384 |
Price history
Qwen3.5-35B-A3B has been in our daily price archive since tracking began on 2026-08-15, when it listed at $0.225 input / $1.80 output per 1M tokens. The price we quote has changed 12 times since then, 11 of them a change of quoted host rather than a repricing. The latest change, on 2026-10-21, was -0.4% on the blended price (30% input, 70% output), to $0.14 input / $1.00 output (quoted host changed, not a repricing). Net of all changes, the blended price is -44.1% against 2026-08-15.
| Date | Old price (in / out) | New price (in / out) | Blended change |
|---|---|---|---|
| 2026-10-21 | $0.15 / $1.00 | $0.14 / $1.00 | -0.4% (quoted host changed, not a repricing) |
| 2026-10-14 | $0.08 / $0.75 | $0.15 / $1.00 | +35.7% (quoted host changed, not a repricing) |
| 2026-10-09 | $0.15 / $1.00 | $0.08 / $0.75 | -26.3% (quoted host changed, not a repricing) |
| 2026-10-07 | $0.08 / $0.75 | $0.15 / $1.00 | +35.7% (quoted host changed, not a repricing) |
| 2026-10-06 | $0.15 / $1.00 | $0.08 / $0.75 | -26.3% (quoted host changed, not a repricing) |
7 earlier changes are not listed.
Alternatives to Qwen3.5-35B-A3B
3 of the 4 comparable models below cost less than Qwen3.5-35B-A3B: Qwen3.5-Flash undercuts it by 73%, and scores 2 Arena points higher.
- Next cheaper model in the Budget price tier: Qwen3 Coder Flash (Alibaba) at $0.741 blended, under 1% less.
- Next more expensive: DeepSeek V3.1 Terminus (DeepSeek) at $0.781 blended, 5% more.
The nearest models in the Budget price tier on Arena ELO, each with its blended price and score difference against Qwen3.5-35B-A3B.
- Qwen3.5-Flash (Alibaba): $0.065 in / $0.26 out per 1M, Arena ELO 1397 (73% cheaper, 2 ELO points higher)
- Claude Haiku 5.5 (Anthropic): $0.10 in / $0.50 out per 1M, Arena ELO 1399 (49% cheaper, 4 ELO points higher)
- Muse Glimmer 30B (meta): $0.30 in / $1.20 out per 1M, Arena ELO 1389 (25% more expensive, 6 ELO points lower)
- GPT-6 Luna (OpenAI): $0.10 in / $0.50 out per 1M, Arena ELO 1388 (49% cheaper, 7 ELO points lower)
- Cheapest within 15 Arena points: Qwen3.5-Flash (Alibaba), 73% cheaper, 2 ELO points higher, out of 21 models in that band.
- Highest Arena ELO within 20% of this price: MiMo-V2.6-Pro (Xiaomi), under 1% cheaper, 97 ELO points higher, out of 18 models in that price band.
Other models from Alibaba
The 8 nearest to Qwen3.5-35B-A3B in price, of 39 other Alibaba models in the catalogue. Prices per 1M tokens, input then output.
- Qwen3 235B A22B Instruct 2507: $0.14 / $0.80
- Qwen-Plus: $0.26 / $0.78
- Qwen3.6 35B A3B: $0.10 / $0.90
- Qwen3 Coder Flash: $0.195 / $0.975
- Qwen3 VL 30B A3B Thinking: $0.29 / $1.00
- Qwen3 Coder 480B A35B: $0.30 / $1.00
- Qwen3 Next 80B A3B Instruct: $0.09 / $1.10
- Qwen3.6 Flash: $0.1875 / $1.125
All 40 Alibaba models and prices
Common questions about Qwen3.5-35B-A3B pricing
- How much does Qwen3.5-35B-A3B cost?
- Qwen3.5-35B-A3B costs $0.14 per 1M input tokens and $1.00 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 $0.1625 per 1M input tokens and $1.30 per 1M output tokens. Prices as listed on 2026-11-12.
- Which host is cheapest for Qwen3.5-35B-A3B?
- Cheapest host: Darkbloom at $0.065 input / $0.75 output per 1M tokens, 27% below the price we quote on the blended price (30% input, 70% output). It serves a quantized fp4 build, which is not the same product as the full-precision model. A support ticket workload at 100K requests a month costs $47.25 there against $71.00 at the price we quote, a difference of $23.75 a month. The dearest, SiliconFlow, charges 2.4x the cheapest. Host rates as listed on 2026-11-12.
- Does Qwen3.5-35B-A3B offer batch or cached pricing?
- Yes, for prompt caching only. Cache reads cost $0.05 per 1M tokens, as listed on 2026-11-12. No batch pricing is published.
- What would Qwen3.5-35B-A3B cost per month?
- With the Support Ticket profile (1,500 input and 500 output tokens per request), 100,000 requests a month cost about $71 on Qwen3.5-35B-A3B at the list price quoted on 2026-11-12.
- Can Qwen3.5-35B-A3B be self-hosted?
- Yes. Qwen3.5-35B-A3B 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 Qwen3.5-35B-A3B changed?
- Qwen3.5-35B-A3B has been in our daily price archive since tracking began on 2026-08-15, when it listed at $0.225 input / $1.80 output per 1M tokens. The price we quote has changed 12 times since then, 11 of them a change of quoted host rather than a repricing. The latest change, on 2026-10-21, was -0.4% on the blended price (30% input, 70% output), to $0.14 input / $1.00 output (quoted host changed, not a repricing). Net of all changes, the blended price is -44.1% 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