Gemini 4 Argon: 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) | 99.88% | 100%◆ (best) | 100%◆ (best) | 3.22s | 18.2s◆ (best) | 60 tps | 1.0M | $2.00 / $10.00 |
Google AI Studiogoogle-ai-studio/flex | Healthy | 100%◆ (best) | 99.63% | 100%◆ (best) | 100%◆ (best) | 6.38s | 74.1s | 34 tps | 1.0M | $1.00 / $5.00◆ (best) |
Google AI Studiogoogle-ai-studio | Down | 40.00% | 80.53% | 39.35% | 39.35% | 5.02s | 80.7s | 44 tps | 1.0M | $2.00 / $10.00 |
Googlegoogle-vertex/global/priority | No data | — | 93.18% | 100%◆ (best) | 100%◆ (best) | 2.79s◆ (best) | 33.1s | 70 tps◆ (best) | 1.0M | $3.60 / $18.00 |
Google AI Studiogoogle-ai-studio/priority | No data | — | 95.65% | — | — | — | — | — | 1.0M | $3.60 / $18.00 |
- Google100%◆ (best)Healthy24h 99.88%
- 7d
- 100%◆ (best)
- 30d
- 100%◆ (best)
- p50
- 3.22s
- p99
- 18.2s◆ (best)
- Throughput
- 60 tps
- Context
- 1.0M
- $/Mtok
- $2.00 / $10.00
google-vertex/global - Google AI Studio100%◆ (best)Healthy24h 99.63%
- 7d
- 100%◆ (best)
- 30d
- 100%◆ (best)
- p50
- 6.38s
- p99
- 74.1s
- Throughput
- 34 tps
- Context
- 1.0M
- $/Mtok
- $1.00 / $5.00◆ (best)
google-ai-studio/flex - Google AI Studio40.00%Down24h 80.53%
- 7d
- 39.35%
- 30d
- 39.35%
- p50
- 5.02s
- p99
- 80.7s
- Throughput
- 44 tps
- Context
- 1.0M
- $/Mtok
- $2.00 / $10.00
google-ai-studio - No data24h 93.18%
- 7d
- 100%◆ (best)
- 30d
- 100%◆ (best)
- p50
- 2.79s◆ (best)
- p99
- 33.1s
- Throughput
- 70 tps◆ (best)
- Context
- 1.0M
- $/Mtok
- $3.60 / $18.00
google-vertex/global/priority - No data24h 95.65%
- 7d
- —
- 30d
- —
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.0M
- $/Mtok
- $3.60 / $18.00
google-ai-studio/priority
Measured 4m 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, averaged over the hour.
Throughput
Median tokens per second.