gpt-oss-20b

openai/gpt-oss-20b

Compare with GPT Audio, GPT Audio Mini, GPT Chat Latest

Comparegpt-oss-20b

	State
Amazon Bedrockamazon-bedrock	Healthy	100%	98.13%	99.06%	99.06%	347ms	4.45s	345 tps	131K	$0.070 / $0.150
DeepInfradeepinfra/bf16bf16	Healthy	100%	99.89%	99.50%	99.50%	311ms	9.34s	79 tps	131K	$0.030 / $0.140
Novitanovita/fp4fp4	Healthy	100%	96.97%	99.13%	99.13%	536ms	30.2s	94 tps	131K	$0.040 / $0.150
SiliconFlowsiliconflow/fp8fp8	Healthy	100%	97.46%	98.07%	98.07%	1.20s	18.7s	51 tps	131K	$0.040 / $0.180
CoreWeavecoreweave/fp4fp4	Healthy	99.63%	99.31%	99.25%	99.25%	278ms	5.46s	90 tps	131K	$0.030 / $0.130
Parasailparasail/fp4fp4	Healthy	99.44%	98.89%	98.77%	98.77%	462ms	4.77s	72 tps	131K	$0.030 / $0.150
Groqgroq	Healthy	99.25%	98.01%	98.12%	98.12%	413ms	2.17s	202 tps	131K	$0.075 / $0.300
Fireworksfireworks	Healthy	98.51%	97.60%	97.20%	97.20%	328ms	4.58s	121 tps	131K	$0.070 / $0.300
Googlegoogle-vertex/us-central1	Down	57.02%	83.14%	79.86%	79.86%	10.4s	26.0s	17 tps	131K	$0.070 / $0.250
Amazon Bedrockamazon-bedrock/eu-west-1	No data	—	—	—	—	—	—	—	131K	$0.070 / $0.150
Phalaphala	Degraded	—	91.87%	92.46%	92.46%	887ms	8.79s	53 tps	131K	$0.040 / $0.150
Togethertogether	No data	—	95.39%	—	—	443ms	1.33s	64 tps	131K	$0.050 / $0.200

Amazon Bedrock100%
Healthy24h 98.13%
7d
99.06%
30d
99.06%
p50
347ms
p99
4.45s
Throughput
345 tps
Context
131K
$/Mtok
$0.070 / $0.150
amazon-bedrock
DeepInfra100%
Healthy24h 99.89%
7d
99.50%
30d
99.50%
p50
311ms
p99
9.34s
Throughput
79 tps
Context
131K
$/Mtok
$0.030 / $0.140
deepinfra/bf16bf16
Novita100%
Healthy24h 96.97%
7d
99.13%
30d
99.13%
p50
536ms
p99
30.2s
Throughput
94 tps
Context
131K
$/Mtok
$0.040 / $0.150
novita/fp4fp4
SiliconFlow100%
Healthy24h 97.46%
7d
98.07%
30d
98.07%
p50
1.20s
p99
18.7s
Throughput
51 tps
Context
131K
$/Mtok
$0.040 / $0.180
siliconflow/fp8fp8
CoreWeave99.63%
Healthy24h 99.31%
7d
99.25%
30d
99.25%
p50
278ms
p99
5.46s
Throughput
90 tps
Context
131K
$/Mtok
$0.030 / $0.130
coreweave/fp4fp4
Parasail99.44%
Healthy24h 98.89%
7d
98.77%
30d
98.77%
p50
462ms
p99
4.77s
Throughput
72 tps
Context
131K
$/Mtok
$0.030 / $0.150
parasail/fp4fp4
Groq99.25%
Healthy24h 98.01%
7d
98.12%
30d
98.12%
p50
413ms
p99
2.17s
Throughput
202 tps
Context
131K
$/Mtok
$0.075 / $0.300
groq
Fireworks98.51%
Healthy24h 97.60%
7d
97.20%
30d
97.20%
p50
328ms
p99
4.58s
Throughput
121 tps
Context
131K
$/Mtok
$0.070 / $0.300
fireworks
Google57.02%
Down24h 83.14%
7d
79.86%
30d
79.86%
p50
10.4s
p99
26.0s
Throughput
17 tps
Context
131K
$/Mtok
$0.070 / $0.250
google-vertex/us-central1
Amazon Bedrock—
No data24h —
7d
—
30d
—
p50
—
p99
—
Throughput
—
Context
131K
$/Mtok
$0.070 / $0.150
amazon-bedrock/eu-west-1
Phala—
Degraded24h 91.87%
7d
92.46%
30d
92.46%
p50
887ms
p99
8.79s
Throughput
53 tps
Context
131K
$/Mtok
$0.040 / $0.150
phala
Together—
No data24h 95.39%
7d
—
30d
—
p50
443ms
p99
1.33s
Throughput
64 tps
Context
131K
$/Mtok
$0.050 / $0.200
together

Measured 3m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.

History

2 endpoints reporting no data are omitted from the charts.

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.