Qwen API pricing
Qwen API prices per million tokens, context windows, open-weights status and independent benchmark scores.
Prices updated . Benchmarks updated .
Prices are the current OpenRouter rate, refreshed about hourly. This page does not chart older prices.
| Model | You send / 1M | It replies / 1M | Context | ECI | Launched |
|---|---|---|---|---|---|
| Qwen3.8 Max (0902)Closed | $2.00 | $6.00 | 984K | 156.2 | 2026-09-02 |
| Qwen3.7 MaxClosed | $1.48 | $4.42 | 1M | 154.2 | 2026-05-21 |
| Qwen3.8 27BOpen weights | $0.040 | $2.30 | 131K | 151.8 | 2026-08-14 |
| Qwen3.6 PlusClosed | $0.325 | $1.95 | 984K | 147.8 | 2026-04-02 |
| Qwen3.7 PlusClosed | $0.320 | $1.28 | 984K | 147.1 | 2026-06-02 |
| Qwen3.5 397B A17BOpen weights | $0.500 | $3.60 | 262K | 146.8 | 2026-02-15 |
| Qwen3.6 27BOpen weights | $0.450 | $2.70 | 16K | 146.3 | 2026-04-22 |
| Qwen3.7 FlashClosed | $0.030 | $0.130 | 984K | 144.1 | 2026-07-15 |
| Qwen3.6 35B A3BOpen weights | $0.100 | $0.900 | 262K | 143.8 | 2026-04-17 |
| Qwen3.6 FlashClosed | $0.188 | $1.13 | 984K | 143.0 | 2026-04-27 |
| Qwen3.5-35B-A3BOpen weights | $0.140 | $1.00 | 262K | 142.6 | 2026-02-23 |
| Qwen3.5-9BOpen weights | $0.170 | $0.250 | 262K | 139.4 | 2026-02-23 |
| Qwen3 32BOpen weights | $0.080 | $0.280 | 41K | 138.5 | 2025-04-28 |
| Qwen3 14BOpen weights | $0.120 | $0.240 | 33K | 138.2 | 2025-04-28 |
| Qwen3 30B A3BOpen weights | $0.120 | $0.500 | 33K | 136.2 | 2025-04-28 |
| Qwen3.8 Max PrimeClosed | $4.00 | $12.00 | 1M | n/a | 2026-09-23 |
| Qwen3.8 Omni FlashClosed | $0.150 | $0.470 | 984K | n/a | 2026-09-17 |
| Qwen3.8 FlashClosed | $0.150 | $0.470 | 1M | n/a | 2026-08-26 |
| Qwen3.8 2.4T A95BOpen weights | $2.00 | $6.00 | 1.01M | n/a | 2026-08-12 |
| Qwen3.5 Plus 2026-04-20Closed | $0.300 | $1.80 | 1M | n/a | 2026-04-27 |
| Qwen3.5-27BOpen weights | $0.260 | $2.60 | 262K | n/a | 2026-02-23 |
| Qwen3.5-122B-A10BOpen weights | $0.260 | $2.08 | 262K | n/a | 2026-02-23 |
| Qwen3.5-FlashClosed | $0.065 | $0.260 | 996K | n/a | 2026-02-23 |
| Qwen3.5 Plus 2026-02-15Closed | $0.260 | $1.56 | 1M | n/a | 2026-02-16 |
| Qwen3 VL 30B A3B ThinkingClosed | $0.290 | $1.00 | 128K | n/a | 2026-02-09 |
| Qwen3 Coder NextOpen weights | $0.120 | $0.800 | 262K | n/a | 2026-02-03 |
| Qwen3 VL 8B InstructOpen weights | $0.250 | $0.750 | 262K | n/a | 2025-10-14 |
| Qwen3 VL 30B A3B InstructOpen weights | $0.150 | $0.600 | 262K | n/a | 2025-10-04 |
| Qwen3 VL 235B A22B InstructOpen weights | $0.210 | $1.90 | 262K | n/a | 2025-09-23 |
| Qwen3 Next 80B A3B InstructOpen weights | $0.090 | $1.10 | 129K | n/a | 2025-09 |
| Qwen3 Coder FlashClosed | $0.195 | $0.975 | 998K | n/a | 2025-07-28 |
| Qwen3 30B A3B Instruct 2507Open weights | $0.100 | $0.300 | 128K | n/a | 2025-07-28 |
| Qwen3 235B A22B Thinking 2507Open weights | $0.450 | $3.50 | 262K | n/a | 2025-07-25 |
| Qwen3 Coder 480B A35BOpen weights | $0.300 | $1.00 | 262K | n/a | 2025-07-22 |
| Qwen3 235B A22B Instruct 2507Open weights | $0.090 | $0.550 | 131K | n/a | 2025-07-21 |
| Qwen3 Coder 30B A3B InstructOpen weights | $0.070 | $0.280 | 262K | n/a | 2025-04 |
| Qwen2.5 VL 72B InstructOpen weights | $0.800 | $1.00 | 128K | n/a | 2025-01-26 |
| Qwen2.5 Coder 32B InstructOpen weights | $0.660 | $1.00 | 33K | n/a | 2024-11-12 |
| Qwen2.5 7B InstructOpen weights | $0.100 | $0.200 | 33K | n/a | 2024-10-16 |
| Qwen2.5 72B InstructOpen weights | $0.360 | $0.400 | 128K | n/a | 2024-09-19 |
| Qwen-PlusClosed | $0.260 | $0.780 | 131K | n/a | 2024-01-25 |
Which Qwen model ranks highest?
Qwen3.8 Max (0902) has the highest Epoch Capabilities Index among Qwen models on this page, at 156.2. Qwen3.7 Max is next at 154.2. Qwen3.8 Max (0902) costs $6.00 per million output tokens and reads 984K of context. The index is an independent score from Epoch AI. Lab-reported numbers are kept off this ranking.
Which Qwen model is cheapest?
Qwen3.7 Flash has the lowest published output price among Qwen models here, at $0.130 per million tokens. Input tokens cost $0.030 per million and the context window is 984K. Qwen2.5 7B Instruct is next at $0.200 per million output tokens. Its Epoch Capabilities Index is 144.1. Prices are the current public rate and can change.
Which Qwen open-weight models lead?
Qwen3.8 27B leads Qwen open-weight models that have a published Epoch Capabilities Index, at 151.8. It is from Qwen, costs $2.30 per million output tokens, and reads 131K. Qwen3.5 397B A17B from Qwen follows at 146.8. Open weights means the lab published the weights. The license on the model page says what you can do with them.
FAQ
Which Qwen model ranks highest?
Qwen3.8 Max (0902) has the highest Epoch Capabilities Index among Qwen models on this page, at 156.2. Qwen3.7 Max is next at 154.2. Qwen3.8 Max (0902) costs $6.00 per million output tokens and reads 984K of context. The index is an independent score from Epoch AI. Lab-reported numbers are kept off this ranking.
Which Qwen model is cheapest?
Qwen3.7 Flash has the lowest published output price among Qwen models here, at $0.130 per million tokens. Input tokens cost $0.030 per million and the context window is 984K. Qwen2.5 7B Instruct is next at $0.200 per million output tokens. Its Epoch Capabilities Index is 144.1. Prices are the current public rate and can change.
How many Qwen models are listed?
This page lists 41 Qwen models with a public price or an independent score. Open the comparison table to filter the full catalogue.
Sources and licenses
- OpenRouter: Prices, context windows, modalities and the model list come from the OpenRouter public API. Data: OpenRouter.
- Vercel AI Gateway: The no-training flag comes from the Vercel AI Gateway model list.
- models.dev: Lab launch dates, open-weights flags and knowledge cutoffs come from models.dev. Copyright (c) 2025 models.dev, MIT License.
- Epoch AI: Independent benchmark scores and the Epoch Capabilities Index (ECI) come from Epoch AI's Benchmarking Hub, used under CC BY 4.0. Epoch AI, Capabilities & benchmarking, epoch.ai/benchmarks. Scores Epoch marks as external keep their original license, and the source is named on each score.
- LMArena: Arena Elo (text, webdev and vision) comes from the LMArena leaderboard dataset (lmarena-ai/leaderboard-dataset), CC BY 4.0, leaderboard date 2026-10-28.
- Lab-reported scores, when an editor has added them, are the lab's own numbers. Each one links to the source and is labeled lab-reported. They are not used to rank the lists.
- What a score measures: the benchmark guide.