Qwen3.8 2.4T A95B is an open-weight sparse mixture-of-experts model from Qwen and the open-weight variant of Qwen3.8 Max, with 95 billion active parameters out of 2.4 trillion total. It is suited for coding, research, complex reasoning, and agentic workflows.
| $2.00 | $6.00 | $0.25 | -- | -- | 4.20s | 31 tps | ||
| $2.00 | $6.00 | $0.25 | $2.50 | $0.17 | 13.03s | 46 tps | ||
| $2.00 | $6.00 | $0.25 | -- | -- | 2.62s | 35 tps | ||
20% off | $2.50$2.00 | $7.50$6.00 | $0.3125$0.25 | -- | -- | 1.11s | 16 tps | |
| $2.00 | $6.00 | $0.25 | -- | -- | 3.08s | 93 tps | ||
 | $2.00 | $6.00 | $0.20 | -- | -- | 1.50s | 80 tps | |
| $2.00 | $6.00 | $0.25 | -- | -- | 0.33s | 227 tps |
P50, best across providers
P50, best provider
When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and customization options.
