Gemma 4 26B A4B : 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 | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Healthy | 100%◆ (best) | 99.97% | 99.96% | 99.56% | 374ms | 2.04s | 54 tps | 262K | $0.100 / $0.300 | |
Darkbloomdarkbloom | Healthy | 100%◆ (best) | 99.99% | 99.99%◆ (best) | 99.94%◆ (best) | 1.01s | 5.13s | 47 tps | 131K | $0.042 / $0.220◆ (best) |
Makoramakora | Healthy | 99.99% | 98.79% | 96.93% | 97.58% | 200ms◆ (best) | 2.09s | 183 tps◆ (best) | 256K | $0.080 / $0.320 |
| Healthy | 99.93% | 99.93% | 99.92% | 99.31% | 242ms | 1.37s◆ (best) | 82 tps | 262K | $0.076 / $0.255 | |
| Healthy | 99.79% | 99.76% | 99.61% | 99.41% | 679ms | 8.70s | 29 tps | 262K | $0.070 / $0.340 | |
| Healthy | 99.53% | 99.80% | 99.79% | 99.49% | 914ms | 4.69s | 44 tps | 262K | $0.130 / $0.400 | |
Cloudflarecloudflare | Healthy | 99.50% | 99.85% | 99.81% | 99.55% | 528ms | 7.06s | 44 tps | 256K | $0.100 / $0.300 |
| Healthy | 99.49% | 99.80% | 99.80% | 99.80% | 606ms | 3.75s | 58 tps | 262K | $0.060 / $0.330 | |
| Healthy | 98.41% | 99.06% | 98.80% | 98.87% | 1.26s | 10.5s | 17 tps | 262K | $0.130 / $0.400 | |
| Healthy | 97.86% | 99.72% | 98.66% | 98.98% | 514ms | 6.08s | 60 tps | 262K | $0.059 / $0.255 | |
| Healthy | 96.73% | 98.73% | 99.42% | 98.94% | 1.36s | 22.7s | 17 tps | 256K | $0.130 / $0.400 | |
Googlegoogle-vertex/global | Degraded | 95.88% | 97.45% | 96.58% | 96.79% | 636ms | 8.50s | 31 tps | 262K | $0.150 / $0.600 |
| Down | 86.29% | 95.71% | 93.56% | 96.43% | 1.55s | 7.40s | 26 tps | 262K | $0.140 / $0.400 | |
Rekareka | No data | — | — | 93.62% | 98.59% | — | — | — | 262K | $0.060 / $0.200 |
- CoreWeave100%◆ (best)Healthy24h 99.97%
- 7d
- 99.96%
- 30d
- 99.56%
- p50
- 374ms
- p99
- 2.04s
- Throughput
- 54 tps
- Context
- 262K
- $/Mtok
- $0.100 / $0.300
coreweave/bf16bf16 - Darkbloom100%◆ (best)Healthy24h 99.99%
- 7d
- 99.99%◆ (best)
- 30d
- 99.94%◆ (best)
- p50
- 1.01s
- p99
- 5.13s
- Throughput
- 47 tps
- Context
- 131K
- $/Mtok
- $0.042 / $0.220◆ (best)
darkbloom - Makora99.99%Healthy24h 98.79%
- 7d
- 96.93%
- 30d
- 97.58%
- p50
- 200ms◆ (best)
- p99
- 2.09s
- Throughput
- 183 tps◆ (best)
- Context
- 256K
- $/Mtok
- $0.080 / $0.320
makora - NextBit99.93%Healthy24h 99.93%
- 7d
- 99.92%
- 30d
- 99.31%
- p50
- 242ms
- p99
- 1.37s◆ (best)
- Throughput
- 82 tps
- Context
- 262K
- $/Mtok
- $0.076 / $0.255
nextbit/bf16bf16 - DeepInfra99.79%Healthy24h 99.76%
- 7d
- 99.61%
- 30d
- 99.41%
- p50
- 679ms
- p99
- 8.70s
- Throughput
- 29 tps
- Context
- 262K
- $/Mtok
- $0.070 / $0.340
deepinfra/fp8fp8 - Novita99.53%Healthy24h 99.80%
- 7d
- 99.79%
- 30d
- 99.49%
- p50
- 914ms
- p99
- 4.69s
- Throughput
- 44 tps
- Context
- 262K
- $/Mtok
- $0.130 / $0.400
novita/bf16bf16 - Cloudflare99.50%Healthy24h 99.85%
- 7d
- 99.81%
- 30d
- 99.55%
- p50
- 528ms
- p99
- 7.06s
- Throughput
- 44 tps
- Context
- 256K
- $/Mtok
- $0.100 / $0.300
cloudflare - DekaLLM99.49%Healthy24h 99.80%
- 7d
- 99.80%
- 30d
- 99.80%
- p50
- 606ms
- p99
- 3.75s
- Throughput
- 58 tps
- Context
- 262K
- $/Mtok
- $0.060 / $0.330
dekallm/bf16bf16 - Parasail98.41%Healthy24h 99.06%
- 7d
- 98.80%
- 30d
- 98.87%
- p50
- 1.26s
- p99
- 10.5s
- Throughput
- 17 tps
- Context
- 262K
- $/Mtok
- $0.130 / $0.400
parasail/bf16bf16 - Io Net97.86%Healthy24h 99.72%
- 7d
- 98.66%
- 30d
- 98.98%
- p50
- 514ms
- p99
- 6.08s
- Throughput
- 60 tps
- Context
- 262K
- $/Mtok
- $0.059 / $0.255
io-net/bf16bf16 - Venice96.73%Healthy24h 98.73%
- 7d
- 99.42%
- 30d
- 98.94%
- p50
- 1.36s
- p99
- 22.7s
- Throughput
- 17 tps
- Context
- 256K
- $/Mtok
- $0.130 / $0.400
venice/bf16bf16 - Google95.88%Degraded24h 97.45%
- 7d
- 96.58%
- 30d
- 96.79%
- p50
- 636ms
- p99
- 8.50s
- Throughput
- 31 tps
- Context
- 262K
- $/Mtok
- $0.150 / $0.600
google-vertex/global - SiliconFlow86.29%Down24h 95.71%
- 7d
- 93.56%
- 30d
- 96.43%
- p50
- 1.55s
- p99
- 7.40s
- Throughput
- 26 tps
- Context
- 262K
- $/Mtok
- $0.140 / $0.400
siliconflow/fp8fp8 - Reka—No data24h —
- 7d
- 93.62%
- 30d
- 98.59%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 262K
- $/Mtok
- $0.060 / $0.200
reka
Measured 6m 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.