gpt-oss-120b

openai/gpt-oss-120b

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

Comparegpt-oss-120b

	State
Amazon Bedrockamazon-bedrock	Healthy	100%	99.07%	99.27%	99.27%	486ms	3.40s	281 tps	131K	$0.150 / $0.600
BaseTenbaseten/fp4fp4	Healthy	100%	99.60%	99.17%	99.17%	194ms	643ms	280 tps	128K	$0.100 / $0.500
Cerebrascerebras/fp16fp16	Healthy	100%	99.99%	99.99%	99.99%	180ms	1.53s	920 tps	131K	$0.350 / $0.750
DeepInfradeepinfra/bf16bf16	Healthy	100%	97.49%	98.23%	98.23%	340ms	5.78s	49 tps	131K	$0.037 / $0.170
DeepInfradeepinfra/turbobf16	Healthy	100%	99.74%	98.93%	98.93%	411ms	1.21s	127 tps	131K	$0.150 / $0.600
Mancer 2mancer/fp8fp8	Healthy	100%	99.46%	99.79%	99.79%	624ms	3.84s	122 tps	131K	$0.085 / $0.500
Parasailparasail/fp4fp4	Healthy	100%	99.66%	99.83%	99.83%	299ms	1.10s	130 tps	131K	$0.100 / $0.750
Phalaphala	Healthy	100%	97.30%	97.42%	97.42%	893ms	3.53s	113 tps	131K	$0.150 / $0.600
SiliconFlowsiliconflow/fp8fp8	Healthy	100%	85.74%	92.47%	92.47%	1.37s	16.3s	20 tps	131K	$0.050 / $0.450
Groqgroq	Healthy	99.97%	99.95%	99.93%	99.93%	226ms	1.91s	364 tps	131K	$0.150 / $0.600
DigitalOceandigitalocean	Healthy	99.91%	99.90%	99.90%	99.90%	843ms	4.51s	25 tps	128K	$0.055 / $0.385
AkashMLakashml/bf16bf16	Healthy	99.76%	99.43%	99.68%	99.68%	933ms	10.3s	69 tps	131K	$0.037 / $0.490
CoreWeavecoreweave/fp4fp4	Healthy	99.71%	94.84%	95.91%	95.91%	424ms	9.32s	35 tps	131K	$0.030 / $0.170
Nebiusnebius/fp4fp4	Healthy	99.47%	99.52%	98.72%	98.72%	218ms	1.47s	283 tps	131K	$0.150 / $0.600
SambaNovasambanova	Healthy	96.49%	88.50%	86.19%	86.19%	835ms	13.8s	395 tps	131K	$0.140 / $0.950
Maramara	Degraded	94.96%	97.71%	95.30%	95.30%	1.41s	10.1s	187 tps	131K	$0.150 / $0.750
Novitanovita/fp4fp4	Degraded	94.46%	92.08%	94.44%	94.44%	610ms	15.8s	47 tps	131K	$0.050 / $0.250
Googlegoogle-vertex/global	Down	87.44%	89.90%	89.82%	89.82%	433ms	9.72s	109 tps	131K	$0.090 / $0.360
Togethertogether	Down	63.44%	96.30%	97.45%	97.45%	484ms	3.39s	117 tps	131K	$0.150 / $0.600
Amazon Bedrockamazon-bedrock/eu-west-1	No data	—	—	100%	100%	—	—	—	131K	$0.150 / $0.600

Amazon Bedrock100%
Healthy24h 99.07%
7d
99.27%
30d
99.27%
p50
486ms
p99
3.40s
Throughput
281 tps
Context
131K
$/Mtok
$0.150 / $0.600
amazon-bedrock
BaseTen100%
Healthy24h 99.60%
7d
99.17%
30d
99.17%
p50
194ms
p99
643ms
Throughput
280 tps
Context
128K
$/Mtok
$0.100 / $0.500
baseten/fp4fp4
Cerebras100%
Healthy24h 99.99%
7d
99.99%
30d
99.99%
p50
180ms
p99
1.53s
Throughput
920 tps
Context
131K
$/Mtok
$0.350 / $0.750
cerebras/fp16fp16
DeepInfra100%
Healthy24h 97.49%
7d
98.23%
30d
98.23%
p50
340ms
p99
5.78s
Throughput
49 tps
Context
131K
$/Mtok
$0.037 / $0.170
deepinfra/bf16bf16
DeepInfra100%
Healthy24h 99.74%
7d
98.93%
30d
98.93%
p50
411ms
p99
1.21s
Throughput
127 tps
Context
131K
$/Mtok
$0.150 / $0.600
deepinfra/turbobf16
Mancer 2100%
Healthy24h 99.46%
7d
99.79%
30d
99.79%
p50
624ms
p99
3.84s
Throughput
122 tps
Context
131K
$/Mtok
$0.085 / $0.500
mancer/fp8fp8
Parasail100%
Healthy24h 99.66%
7d
99.83%
30d
99.83%
p50
299ms
p99
1.10s
Throughput
130 tps
Context
131K
$/Mtok
$0.100 / $0.750
parasail/fp4fp4
Phala100%
Healthy24h 97.30%
7d
97.42%
30d
97.42%
p50
893ms
p99
3.53s
Throughput
113 tps
Context
131K
$/Mtok
$0.150 / $0.600
phala
SiliconFlow100%
Healthy24h 85.74%
7d
92.47%
30d
92.47%
p50
1.37s
p99
16.3s
Throughput
20 tps
Context
131K
$/Mtok
$0.050 / $0.450
siliconflow/fp8fp8
Groq99.97%
Healthy24h 99.95%
7d
99.93%
30d
99.93%
p50
226ms
p99
1.91s
Throughput
364 tps
Context
131K
$/Mtok
$0.150 / $0.600
groq
DigitalOcean99.91%
Healthy24h 99.90%
7d
99.90%
30d
99.90%
p50
843ms
p99
4.51s
Throughput
25 tps
Context
128K
$/Mtok
$0.055 / $0.385
digitalocean
AkashML99.76%
Healthy24h 99.43%
7d
99.68%
30d
99.68%
p50
933ms
p99
10.3s
Throughput
69 tps
Context
131K
$/Mtok
$0.037 / $0.490
akashml/bf16bf16
CoreWeave99.71%
Healthy24h 94.84%
7d
95.91%
30d
95.91%
p50
424ms
p99
9.32s
Throughput
35 tps
Context
131K
$/Mtok
$0.030 / $0.170
coreweave/fp4fp4
Nebius99.47%
Healthy24h 99.52%
7d
98.72%
30d
98.72%
p50
218ms
p99
1.47s
Throughput
283 tps
Context
131K
$/Mtok
$0.150 / $0.600
nebius/fp4fp4
SambaNova96.49%
Healthy24h 88.50%
7d
86.19%
30d
86.19%
p50
835ms
p99
13.8s
Throughput
395 tps
Context
131K
$/Mtok
$0.140 / $0.950
sambanova
Mara94.96%
Degraded24h 97.71%
7d
95.30%
30d
95.30%
p50
1.41s
p99
10.1s
Throughput
187 tps
Context
131K
$/Mtok
$0.150 / $0.750
mara
Novita94.46%
Degraded24h 92.08%
7d
94.44%
30d
94.44%
p50
610ms
p99
15.8s
Throughput
47 tps
Context
131K
$/Mtok
$0.050 / $0.250
novita/fp4fp4
Google87.44%
Down24h 89.90%
7d
89.82%
30d
89.82%
p50
433ms
p99
9.72s
Throughput
109 tps
Context
131K
$/Mtok
$0.090 / $0.360
google-vertex/global
Together63.44%
Down24h 96.30%
7d
97.45%
30d
97.45%
p50
484ms
p99
3.39s
Throughput
117 tps
Context
131K
$/Mtok
$0.150 / $0.600
together
Amazon Bedrock—
No data24h —
7d
100%
30d
100%
p50
—
p99
—
Throughput
—
Context
131K
$/Mtok
$0.150 / $0.600
amazon-bedrock/eu-west-1

Measured 4m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.

Price changes

gpt-oss-120b Mancer 2mancer/fp8$0.080 / $0.500 → $0.085 / $0.5002026-08-20 23:41 UTC
gpt-oss-120b Mancer 2mancer/fp8$0.083 / $0.500 → $0.080 / $0.5002026-08-20 22:26 UTC
gpt-oss-120b Mancer 2mancer/fp8$0.085 / $0.500 → $0.083 / $0.5002026-08-20 21:56 UTC
gpt-oss-120b Mancer 2mancer/fp8$0.080 / $0.500 → $0.085 / $0.5002026-08-20 21:21 UTC
gpt-oss-120b Mancer 2mancer/fp8$0.085 / $0.500 → $0.080 / $0.5002026-08-19 19:25 UTC

History

1 endpoint reporting no data is 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.