Google: Gemma 4 31B Performance
Compare measured TTFT, throughput, uptime, and route health for Google: Gemma 4 31B across TrustedRouter providers using metadata-only production probes.
google/gemma-4-31b-it
Measured performance
Continuously sampled p50/p95 time-to-first-token (TTFT), effective throughput, and success rate for Google: Gemma 4 31B. 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 |
|---|---|---|---|---|---|---|
| tinfoil | 979 ms | 2343 ms | 255 tok/s n=1 | 100.00% | — | 4 |
| friendli | 1217 ms | 3666 ms | 321 tok/s n=1 | 100.00% | — | 6 |
| sail-research | 1691 ms | 2073 ms | — | 100.00% | — | 2 |
| siliconflow | 2154 ms | 2154 ms | 17 tok/s n=1 | 100.00% | — | 6 |
| regolo | 2979 ms | 2979 ms | 126 tok/s n=1 | 100.00% | — | 1 |
| wandb | 3910 ms | 3910 ms | 51 tok/s n=1 | 100.00% | — | 1 |
| sambanova | 5405 ms | 19308 ms | — | 85.71% | — | 7 |
| novita | — | — | 99 tok/s n=1 | — | — | 0 |
| nvidia-nim | — | — | 5 tok/s n=1 | — | — | 0 |
Full provider & model leaderboard.
14 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.