Qwen3.5 397B A17B
by Qwen · qwen/qwen3.5-397b-a17b
The Qwen3.5 series 397B-A17B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. It delivers...
- Input / 1M tokens
- $0.50
- Output / 1M tokens
- $3.60
- Context
- 262K
Specs
| Input price | $0.50 / 1M tokens |
|---|---|
| Output price | $3.60 / 1M tokens |
| Cached input | $0.30 / 1M tokens |
| Context window | 262K tokens |
| Max output | 236K tokens |
| Reads | text, image, video |
| Writes | text |
| Tool calling | Yes |
| Reasoning | Yes |
| Listed | Feb 16, 2026 |
List prices in USD, updated Oct 30, 2026. Providers may charge differently.
Providers
10 providers serve this model, as of Oct 30, 2026, 18:12 UTC.
| Provider | Quantization | Context | Max output | Input / 1M | Output / 1M | Cache read / 1M | Uptime (30 min) | Latency (median) | Throughput (median) | Supports | Status |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Alibaba | — | 262K | 65.5K | $0.39 | $2.34 | — | 100% | 1.23 s | 52 tok/s | Normal | |
| DeepInfra | fp8 | 262K | 81.9K | $0.45 | $3.00 | $0.22 | 100% | 820 ms | 53 tok/s | Normal | |
| AtlasCloud | fp8 | 262K | 65.5K | $0.55 | $3.50 | $0.55 | 33.3% | 973 ms | 47 tok/s | Code -2 | |
| DigitalOcean | — | 131K | 118K | $0.55 | $3.50 | $0.11 | — | 870 ms | 24 tok/s | Normal | |
| Phala | — | 262K | 236K | $0.55 | $3.50 | $0.225 | — | 2.32 s | 52 tok/s | Normal | |
| Parasail | fp8 | 262K | 236K | $0.50 | $3.60 | $0.30 | 100% | 1.44 s | 73 tok/s | Normal | |
| GMICloud | fp8 | 262K | 236K | $0.60 | $3.60 | — | — | 2.70 s | 42 tok/s | Normal | |
| Novita | — | 262K | 65.5K | $0.60 | $3.60 | — | 100% | 1.32 s | 44 tok/s | Normal | |
| StreamLake | — | 256K | 64K | $0.60 | $3.60 | $0.12 | 100% | 951 ms | 157 tok/s | Normal | |
| Venice | — | 128K | 32.8K | $0.75 | $4.50 | — | — | 3.90 s | 55 tok/s | Normal |
"Cheapest" is the lowest input plus output price; "Fastest" the highest median throughput. Columns empty for every provider are not shown.
Related models
- Qwen
Qwen3.8 Flash
- Input
- $0.15
- Output
- $0.47
- Context
- 1M
- Qwen
Qwen3.8 27B
- Input
- $0.15
- Output
- $1.88
- Context
- 1M
- Qwen
Qwen3.7 Flash
- Input
- $0.03
- Output
- $0.13
- Context
- 1M
- Qwen
Qwen3.8 Max Prime
- Input
- $4.00
- Output
- $12.00
- Context
- 1M
- Qwen
Qwen3.8 Omni Flash
- Input
- $0.15
- Output
- $0.47
- Context
- 1M
- Qwen
Qwen3.8 Max (0902)
- Input
- $2.00
- Output
- $6.00
- Context
- 1M