Gemini 3.8 Flash: providers compared
Every endpoint serving this model, side by side. ◆ marks the best value in each column — per model, never fleet-wide.
Current state
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Googlegoogle-vertex/global | Healthy | 100%◆ (best) | 100% | 99.56% | 97.36% | 2.80s | 29.9s | 69 tps | 1.0M | $0.750 / $3.75 |
Googlegoogle-vertex/us | Healthy | 100%◆ (best) | 99.50% | 93.51% | 94.41% | 3.06s | 45.3s | 94 tps◆ (best) | 1.0M | $0.825 / $4.13 |
Google AI Studiogoogle-ai-studio | Healthy | 100%◆ (best) | 100% | 100%◆ (best) | 99.98%◆ (best) | 2.72s | 13.1s | 73 tps | 1.0M | $0.750 / $3.75 |
Google AI Studiogoogle-ai-studio/flex | Healthy | 100%◆ (best) | 100% | 100%◆ (best) | 99.97% | 2.77s | 11.2s | 75 tps | 1.0M | $0.375 / $1.88◆ (best) |
Google AI Studiogoogle-ai-studio/priority | Healthy | 100%◆ (best) | 100% | 99.73% | 99.54% | 3.44s | 8.24s◆ (best) | 81 tps | 1.0M | $1.35 / $6.75 |
Googlegoogle-vertex/global/flex | Down | — | 71.91% | 86.43% | 86.65% | 37.9s | 222s | 19 tps | 1.0M | $0.375 / $1.88◆ (best) |
Googlegoogle-vertex/global/priority | No data | — | 100% | 99.78% | 99.15% | 2.10s◆ (best) | 13.1s | 33 tps | 1.0M | $1.35 / $6.75 |
- Google100%◆ (best)Healthy24h 100%
- 7d
- 99.56%
- 30d
- 97.36%
- p50
- 2.80s
- p99
- 29.9s
- Throughput
- 69 tps
- Context
- 1.0M
- $/Mtok
- $0.750 / $3.75
google-vertex/global - Google100%◆ (best)Healthy24h 99.50%
- 7d
- 93.51%
- 30d
- 94.41%
- p50
- 3.06s
- p99
- 45.3s
- Throughput
- 94 tps◆ (best)
- Context
- 1.0M
- $/Mtok
- $0.825 / $4.13
google-vertex/us - Google AI Studio100%◆ (best)Healthy24h 100%
- 7d
- 100%◆ (best)
- 30d
- 99.98%◆ (best)
- p50
- 2.72s
- p99
- 13.1s
- Throughput
- 73 tps
- Context
- 1.0M
- $/Mtok
- $0.750 / $3.75
google-ai-studio - Google AI Studio100%◆ (best)Healthy24h 100%
- 7d
- 100%◆ (best)
- 30d
- 99.97%
- p50
- 2.77s
- p99
- 11.2s
- Throughput
- 75 tps
- Context
- 1.0M
- $/Mtok
- $0.375 / $1.88◆ (best)
google-ai-studio/flex - Google AI Studio100%◆ (best)Healthy24h 100%
- 7d
- 99.73%
- 30d
- 99.54%
- p50
- 3.44s
- p99
- 8.24s◆ (best)
- Throughput
- 81 tps
- Context
- 1.0M
- $/Mtok
- $1.35 / $6.75
google-ai-studio/priority - Down24h 71.91%
- 7d
- 86.43%
- 30d
- 86.65%
- p50
- 37.9s
- p99
- 222s
- Throughput
- 19 tps
- Context
- 1.0M
- $/Mtok
- $0.375 / $1.88◆ (best)
google-vertex/global/flex - No data24h 100%
- 7d
- 99.78%
- 30d
- 99.15%
- p50
- 2.10s◆ (best)
- p99
- 13.1s
- Throughput
- 33 tps
- Context
- 1.0M
- $/Mtok
- $1.35 / $6.75
google-vertex/global/priority
Measured 1m ago. Latency and throughput are OpenRouter's rolling 30-minute windows. ◆ marks the best value in each column among these endpoints.
Price vs latency
One dot per endpoint at the latest round · further left = faster, lower = cheaper · hover a dot for its tag
History
Availability
Axis starts at 90% so small differences stay visible; it expands to 0 when an endpoint falls below.
Time to first token
Logarithmic axis — the fleet spans two orders of magnitude. P50 is OpenRouter's rolling 30-minute window.
Throughput
Median tokens per second.