Gemma 4 26B A4B
google/gemma-4-26b-a4b-it
Compare with Gemma 4 31B, Gemma 4 26B A4B (free), Gemma 4 31B (free)
CompareGemma 4 26B A4B
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Healthy | 100% | 93.34% | 95.33% | 95.33% | 878ms | 5.59s | 32 tps | 262K | $0.130 / $0.400 | |
| Healthy | 100% | 99.17% | 99.32% | 99.32% | 1.30s | 5.91s | 31 tps | 262K | $0.120 / $0.400 | |
| Healthy | 99.95% | 99.81% | 99.41% | 99.41% | 365ms | 2.94s | 27 tps | 262K | $0.070 / $0.340 | |
| Healthy | 99.93% | 99.72% | 99.82% | 99.82% | 654ms | 2.46s | 53 tps | 262K | $0.100 / $0.400 | |
Cloudflarecloudflare | Healthy | 99.92% | 97.08% | 94.01% | 94.01% | 561ms | 11.6s | 51 tps | 256K | $0.100 / $0.300 |
| Healthy | 99.72% | 99.34% | 99.32% | 99.32% | 995ms | 7.86s | 29 tps | 256K | $0.130 / $0.400 | |
| Healthy | 99.52% | 98.35% | 98.57% | 98.57% | 532ms | 6.71s | 29 tps | 262K | $0.130 / $0.400 | |
Googlegoogle-vertex/global | Healthy | 99.43% | 97.50% | 98.01% | 98.01% | 374ms | 10.1s | 19 tps | 262K | $0.150 / $0.600 |
Darkbloomdarkbloom | Degraded | 91.90% | 96.00% | 96.19% | 96.19% | 4.20s | 27.1s | 5.0 tps | 131K | $0.050 / $0.250 |
- Novita100%Healthy24h 93.34%
- 7d
- 95.33%
- 30d
- 95.33%
- p50
- 878ms
- p99
- 5.59s
- Throughput
- 32 tps
- Context
- 262K
- $/Mtok
- $0.130 / $0.400
novita/bf16bf16 - SiliconFlow100%Healthy24h 99.17%
- 7d
- 99.32%
- 30d
- 99.32%
- p50
- 1.30s
- p99
- 5.91s
- Throughput
- 31 tps
- Context
- 262K
- $/Mtok
- $0.120 / $0.400
siliconflow/fp8fp8 - DeepInfra99.95%Healthy24h 99.81%
- 7d
- 99.41%
- 30d
- 99.41%
- p50
- 365ms
- p99
- 2.94s
- Throughput
- 27 tps
- Context
- 262K
- $/Mtok
- $0.070 / $0.340
deepinfra/fp8fp8 - NextBit99.93%Healthy24h 99.72%
- 7d
- 99.82%
- 30d
- 99.82%
- p50
- 654ms
- p99
- 2.46s
- Throughput
- 53 tps
- Context
- 262K
- $/Mtok
- $0.100 / $0.400
nextbit/bf16bf16 - Cloudflare99.92%Healthy24h 97.08%
- 7d
- 94.01%
- 30d
- 94.01%
- p50
- 561ms
- p99
- 11.6s
- Throughput
- 51 tps
- Context
- 256K
- $/Mtok
- $0.100 / $0.300
cloudflare - Venice99.72%Healthy24h 99.34%
- 7d
- 99.32%
- 30d
- 99.32%
- p50
- 995ms
- p99
- 7.86s
- Throughput
- 29 tps
- Context
- 256K
- $/Mtok
- $0.130 / $0.400
venice/bf16bf16 - Parasail99.52%Healthy24h 98.35%
- 7d
- 98.57%
- 30d
- 98.57%
- p50
- 532ms
- p99
- 6.71s
- Throughput
- 29 tps
- Context
- 262K
- $/Mtok
- $0.130 / $0.400
parasail/bf16bf16 - Google99.43%Healthy24h 97.50%
- 7d
- 98.01%
- 30d
- 98.01%
- p50
- 374ms
- p99
- 10.1s
- Throughput
- 19 tps
- Context
- 262K
- $/Mtok
- $0.150 / $0.600
google-vertex/global - Darkbloom91.90%Degraded24h 96.00%
- 7d
- 96.19%
- 30d
- 96.19%
- p50
- 4.20s
- p99
- 27.1s
- Throughput
- 5.0 tps
- Context
- 131K
- $/Mtok
- $0.050 / $0.250
darkbloom
Measured 1m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.
Price changes
- Gemma 4 26B A4B Darkbloomdarkbloom$0.030 / $0.165 → $0.050 / $0.2502026-08-20 20:05 UTC
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.