Gemma 4 26B A4B

google/gemma-4-26b-a4b-itInputOutput

Compare its 11 providers Incident RSSCompare with Gemma 4 31B, Gemma 4 26B A4B (free), Gemma 4 31B (free)

CompareGemma 4 26B A4B

	State
Darkbloomdarkbloom	Healthy	99.99%	99.89%	99.82%	98.34%	3.71s	11.6s	9.0 tps	131K	$0.042 / $0.220
Cloudflarecloudflare	Healthy	99.94%	99.91%	99.84%	99.32%	312ms	11.0s	29 tps	256K	$0.100 / $0.300
DekaLLMdekallm/bf16bf16	Healthy	99.87%	99.94%	99.56%	99.56%	428ms	3.82s	80 tps	262K	$0.060 / $0.330
Parasailparasail/bf16bf16	Healthy	99.57%	98.08%	98.66%	98.41%	1.41s	6.03s	28 tps	262K	$0.130 / $0.400
NextBitnextbit/bf16bf16	Healthy	98.79%	97.84%	99.63%	99.50%	604ms	5.39s	29 tps	262K	$0.090 / $0.300
SiliconFlowsiliconflow/fp8fp8	Healthy	98.43%	96.57%	97.58%	98.48%	1.86s	9.08s	21 tps	262K	$0.140 / $0.400
Makoramakora	Healthy	97.12%	98.41%	95.96%	95.96%	780ms	13.2s	49 tps	262K	$0.100 / $0.340
Googlegoogle-vertex/global	Degraded	96.10%	97.58%	95.86%	97.80%	1.11s	13.4s	27 tps	262K	$0.150 / $0.600
Novitanovita/bf16bf16	Degraded	94.75%	98.11%	99.06%	98.48%	955ms	20.2s	30 tps	262K	$0.130 / $0.400
Venicevenice/bf16bf16	Degraded	93.72%	96.50%	98.23%	98.68%	1.75s	16.5s	16 tps	256K	$0.130 / $0.400
DeepInfradeepinfra/fp8fp8	Down	89.29%	98.92%	98.76%	99.45%	4.66s	235s	2.0 tps	262K	$0.070 / $0.340

Darkbloom99.99%
Healthy24h 99.89%
7d
99.82%
30d
98.34%
p50
3.71s
p99
11.6s
Throughput
9.0 tps
Context
131K
$/Mtok
$0.042 / $0.220
darkbloom
Cloudflare99.94%
Healthy24h 99.91%
7d
99.84%
30d
99.32%
p50
312ms
p99
11.0s
Throughput
29 tps
Context
256K
$/Mtok
$0.100 / $0.300
cloudflare
DekaLLM99.87%
Healthy24h 99.94%
7d
99.56%
30d
99.56%
p50
428ms
p99
3.82s
Throughput
80 tps
Context
262K
$/Mtok
$0.060 / $0.330
dekallm/bf16bf16
Parasail99.57%
Healthy24h 98.08%
7d
98.66%
30d
98.41%
p50
1.41s
p99
6.03s
Throughput
28 tps
Context
262K
$/Mtok
$0.130 / $0.400
parasail/bf16bf16
NextBit98.79%
Healthy24h 97.84%
7d
99.63%
30d
99.50%
p50
604ms
p99
5.39s
Throughput
29 tps
Context
262K
$/Mtok
$0.090 / $0.300
nextbit/bf16bf16
SiliconFlow98.43%
Healthy24h 96.57%
7d
97.58%
30d
98.48%
p50
1.86s
p99
9.08s
Throughput
21 tps
Context
262K
$/Mtok
$0.140 / $0.400
siliconflow/fp8fp8
Makora97.12%
Healthy24h 98.41%
7d
95.96%
30d
95.96%
p50
780ms
p99
13.2s
Throughput
49 tps
Context
262K
$/Mtok
$0.100 / $0.340
makora
Google96.10%
Degraded24h 97.58%
7d
95.86%
30d
97.80%
p50
1.11s
p99
13.4s
Throughput
27 tps
Context
262K
$/Mtok
$0.150 / $0.600
google-vertex/global
Novita94.75%
Degraded24h 98.11%
7d
99.06%
30d
98.48%
p50
955ms
p99
20.2s
Throughput
30 tps
Context
262K
$/Mtok
$0.130 / $0.400
novita/bf16bf16
Venice93.72%
Degraded24h 96.50%
7d
98.23%
30d
98.68%
p50
1.75s
p99
16.5s
Throughput
16 tps
Context
256K
$/Mtok
$0.130 / $0.400
venice/bf16bf16
DeepInfra89.29%
Down24h 98.92%
7d
98.76%
30d
99.45%
p50
4.66s
p99
235s
Throughput
2.0 tps
Context
262K
$/Mtok
$0.070 / $0.340
deepinfra/fp8fp8

Measured 2m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.

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.

Daily availability

One row per endpoint · one dot per day · last 90 days.

Jun

Jul

Aug

Sep

worst hour ≥ 95%90–95%< 90%no data

Colour by each endpoint's worst hour of each UTC day · 11 endpoints · 30 days with data, 60 endpoint-days below 90%.

Price changes

Gemma 4 26B A4B NextBitnextbit/bf16$0.100 → $0.090 / $0.400 → $0.3002026-09-11 09:40 UTC
Gemma 4 26B A4B SiliconFlowsiliconflow/fp8$0.120 → $0.140 / $0.400cache read — → $0.0502026-09-10 05:45 UTC
Gemma 4 26B A4B Darkbloomdarkbloom$0.050 → $0.042 / $0.250 → $0.2202026-08-23 01:55 UTC
Gemma 4 26B A4B Darkbloomdarkbloom$0.030 → $0.050 / $0.165 → $0.2502026-08-20 20:05 UTC