Qwen API pricing, every Alibaba model per million tokens
On 10 November 2026, Alibaba models on this list cost from $0.011 per million input tokens (Qwen3.6 35B A3B) to $4.00 (Qwen3.8 Max Prime), and from $0.101 per million output tokens (Qwen3.6 35B A3B) to $12.00 (Qwen3.8 Max Prime).
| Model | Input | Output | Cached input | Batch in / out | Context | Your month | Listed |
|---|---|---|---|---|---|---|---|
| Qwen3.8 Max Prime | $4.00 | $12.00 | $0.50 | 1M | $420.00 | 2026-09-23 | |
| Qwen3.8 Omni Flash | $0.15 | $0.47 | $0.016 | 1M | $16.05 | 2026-09-21 | |
| Qwen3.8 Max (0902) | $2.00 | $6.00 | $0.25 | $1.15 / $3.45 | 1M | $210.00 | 2026-09-03 |
| Qwen3.8 Flash | $0.15 | $0.47 | $0.016 | 1M | $16.05 | 2026-08-26 | |
| Qwen3.8 27B | $0.425 | $2.55 | $0.085 | $0.16 / $2.00 | 1.05M | $63.75 | 2026-08-14 |
| Qwen3.8 2.4T A95B | $2.00 | $6.00 | $0.25 | 1.05M | $210.00 | 2026-08-12 | |
| Qwen3.7 Flash | $0.03 | $0.13 | $0.006 | 1M | $3.75 | 2026-07-27 | |
| Qwen3.7 Plus | $0.32 | $1.28 | $0.064 | $0.23 / $0.92 | 1M | $38.40 | 2026-06-03 |
| Qwen3.7 Max | $1.48 | $4.42 | $0.295 | 1M | $154.88 | 2026-05-21 | |
| Qwen3.5 Plus 2026-04-20 | $0.30 | $1.80 | 1M | $45.00 | 2026-04-27 | ||
| Qwen3.6 Flash | $0.1875 | $1.12 | 1M | $28.12 | 2026-04-27 | ||
| Qwen3.6 35B A3B PrimeIntellect, cheapest of 11 hosts | $0.011 | $0.101 | $0.01 | 262K | $2.17 | 2026-04-27 | |
| Qwen3.6 27B | $0.45 | $2.70 | 262K | $67.50 | 2026-04-27 | ||
| Qwen3.6 Plus | $0.325 | $1.95 | 1M | $48.75 | 2026-04-02 | ||
| Qwen3.5-9B SiliconFlow, cheapest of 6 hosts | $0.10 | $0.15 | 262K | $8.25 | 2026-03-10 | ||
| Qwen3.5-35B-A3B | $0.1625 | $1.30 | 262K | $29.25 | 2026-02-25 | ||
| Qwen3.5-27B | $0.195 | $1.56 | 262K | $35.10 | 2026-02-25 | ||
| Qwen3.5-122B-A10B | $0.26 | $2.08 | 262K | $46.80 | 2026-02-25 | ||
| Qwen3.5-Flash | $0.065 | $0.26 | 1M | $7.80 | 2026-02-25 | ||
| Qwen3.5 Plus 2026-02-15 | $0.26 | $1.56 | 1M | $39.00 | 2026-02-16 | ||
| Qwen3.5 397B A17B | $0.39 | $2.34 | 262K | $58.50 | 2026-02-16 | ||
| Qwen3 Coder Next served by Parasail | $0.12 | $0.80 | $0.07 | 262K | $19.20 | 2026-02-04 | |
| Qwen3 VL 8B Instruct served by Parasail | $0.25 | $0.75 | $0.12 | 262K | $26.25 | 2025-10-14 | |
| Qwen3 VL 30B A3B Thinking served by SiliconFlow | $0.29 | $1.00 | 262K | $32.40 | 2025-10-06 | ||
| Qwen3 VL 30B A3B Instruct DeepInfra, cheapest of 3 hosts | $0.15 | $0.60 | 262K | $18.00 | 2025-10-06 | ||
| Qwen3 VL 235B A22B Instruct DeepInfra, cheapest of 3 hosts | $0.20 | $0.88 | $0.11 | 262K | $25.20 | 2025-09-23 | |
| Qwen3 Coder Flash | $0.195 | $0.975 | $0.039 | 1M | $26.32 | 2025-09-17 | |
| Qwen3 Next 80B A3B Instruct DeepInfra, cheapest of 2 hosts | $0.09 | $1.10 | 262K | $21.90 | 2025-09-11 | ||
| Qwen3 Coder 30B A3B Instruct SiliconFlow, cheapest of 2 hosts | $0.07 | $0.28 | 262K | $8.40 | 2025-07-31 | ||
| Qwen3 30B A3B Instruct 2507 DekaLLM, cheapest of 3 hosts | $0.09 | $0.30 | 262K | $9.90 | 2025-07-29 | ||
| Qwen3 235B A22B Thinking 2507 served by Venice | $0.45 | $3.50 | 128K | $79.50 | 2025-07-25 | ||
| Qwen3 Coder 480B A35B DeepInfra, cheapest of 2 hosts | $0.30 | $1.00 | $0.10 | 262K | $33.00 | 2025-07-23 | |
| Qwen3 235B A22B Instruct 2507 DeepInfra, cheapest of 4 hosts | $0.09 | $0.55 | 262K | $13.65 | 2025-07-21 | ||
| Qwen3 30B A3B served by DeepInfra | $0.12 | $0.50 | 41K | $14.70 | 2025-04-28 | ||
| Qwen3 14B NextBit, cheapest of 2 hosts | $0.10 | $0.22 | 41K | $9.30 | 2025-04-28 | ||
| Qwen3 32B DeepInfra, cheapest of 3 hosts | $0.08 | $0.28 | 131K | $9.00 | 2025-04-28 | ||
| Qwen2.5 VL 72B Instruct served by Parasail | $0.80 | $1.00 | $0.40 | 128K | $63.00 | 2025-02-01 | |
| Qwen-Plus | $0.26 | $0.78 | $0.052 | 1M | $27.30 | 2025-02-01 | |
| Qwen2.5 Coder 32B Instruct served by Cloudflare | $0.66 | $1.00 | 33K | $54.60 | 2024-11-11 | ||
| Qwen2.5 7B Instruct served by Phala | $0.10 | $0.20 | 33K | $9.00 | 2024-10-16 | ||
| Qwen2.5 72B Instruct served by DeepInfra | $0.36 | $0.40 | 33K | $27.60 | 2024-09-19 |
20 of these 41 models are not served by Alibaba itself on OpenRouter, so their price is the cheapest regular price among the hosts that run them, and the host is named under the model. Compare Alibaba with every other provider
Common questions
How much does the Qwen API cost?
On 10 November 2026, Alibaba models on this list cost from $0.011 per million input tokens (Qwen3.6 35B A3B) to $4.00 (Qwen3.8 Max Prime), and from $0.101 per million output tokens (Qwen3.6 35B A3B) to $12.00 (Qwen3.8 Max Prime).
What is the cheapest Qwen model?
For a typical mix of three input tokens to every output token, the cheapest Alibaba model on this list is Qwen3.6 35B A3B at $0.011 per million input tokens and $0.101 per million output tokens.
What is the newest Qwen model on the list?
Qwen3.8 Max Prime, listed on 23 September 2026, at $4.00 per million input tokens and $12.00 per million output tokens.
Which Qwen model has the largest context window?
Qwen3.8 2.4T A95B, with 1.05M tokens of context.
Does Alibaba offer batch pricing?
Yes, 3 of its models on this list show a batch price, with batch input at a median of 57 percent of the standard price. Batch suits work that can wait for its answer.
Choosing a model for a production agent
Price is one input. Quality on your own cases, latency and the cost of retries decide the rest. Tell me what you are building.