Qwen: Qwen3.5 397B A17B
qwen/qwen3-5-397b-a17bThe 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.
Providers
10 providers serve this model upstream.
| Provider | Input /M | Output /M | Cache read /M | Latency P50 | Throughput P50 | Uptime |
|---|---|---|---|---|---|---|
| $0.50 | $3.60 | $0.30 | 0.46s | 84 tps | 99.97% | |
| $0.60 | $3.60 | Variable | 1.41s | 45 tps | 99.92% | |
| $0.39 | $2.34 | Variable | 1.52s | 49 tps | 99.81% | |
 DeepInfraus · fp8 | $0.45 | $3.00 | $0.22 | 0.61s | 42 tps | 99.69% |
| $0.55 | $3.50 | $0.23 | 2.69s | 48 tps | 99.55% | |
| $0.60 | $3.60 | $0.12 | 0.49s | 162 tps | 97.98% | |
| $0.60 | $3.60 | Variable | 1.69s | 57 tps | 91.03% | |
 DigitalOcean | $0.55 | $3.50 | $0.11 | 1.38s | 12 tps | 90.25% |
| $0.75 | $4.50 | Variable | 1.15s | 47 tps | 85.71% | |
| $0.55 | $3.50 | $0.55 | 1.53s | 57 tps | 81% |
Market-wide measurements, refreshed every 15 minutes. They describe who serves Qwen3.5 397B A17B in general, not how ProjectCOZY routes your requests.
Pricing
List rates per 1M tokens — what Qwen3.5 397B A17B costs pay-as-you-go elsewhere, not what you pay here.
Performance
Time to first token and output speed across every provider, over the last 3 days.
Uptime
100% of requests were served successfully over the last 3 days, across all providers.
Apps
Tools that most often run Qwen3.5 397B A17B, by tokens this week.
Activity
191.4B tokens processed market-wide in the last month.
FAQ
Explore
Other models your pass can call.
