DeepSeek: DeepSeek V4 Pro 0423 Performance
Try DeepSeek V4 Pro 0423 on us. Starter credit covers your first requests, then one key reaches all 15 provider routes with automatic failover.
deepseek/deepseek-v4-pro
Measured performance
Continuously sampled p50/p95 time-to-first-token (TTFT), effective throughput, and success rate for DeepSeek: DeepSeek V4 Pro 0423. Effective throughput uses provider-reported output tokens over complete request time. Unsupported route and probe-configuration rows are separated from provider downtime, and no prompt or output content is stored.
| Provider | p50 TTFT | p95 TTFT | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|---|
| deepseek | 1638 ms | 2620 ms | — | 100.00% | — | 13 |
| wafer | 812 ms | 954 ms | — | 100.00% | — | 2 |
| gmi | 1844 ms | 2103 ms | — | 100.00% | — | 2 |
| io-net | 2046 ms | 3486 ms | — | 100.00% | — | 3 |
| novita | 2601 ms | 2811 ms | — | 100.00% | — | 3 |
| digitalocean | 3601 ms | 3981 ms | — | 100.00% | — | 3 |
| siliconflow | 3977 ms | 3977 ms | — | 100.00% | — | 1 |
| featherless | 4600 ms | 4600 ms | — | 100.00% | — | 1 |
Full provider & model leaderboard.
15 routes.
More routes give the auto router more room to fail over around provider 429 and 5xx responses.
Gateway overhead is measured separately.
Public status separates TLS/health overhead from full model latency so slow LLMs do not inflate the router metric.
Metadata rollups.
Status samples store latency, outcome, provider, model, route, cost, and region metadata only.
View public status or inspect provider routes.