Skip to content
ProviderBench home

Cheapest gpt-oss-20b Inference Providers

Ranks providers by estimated cost using a 1,000-input/500-output token ratio, shown per 1 million total tokens.

Weights & Biases currently ranks first at $0.063333 / 1M tokens via wandb/fp4.

Provider options:
13
Ranked:
13
Metric:
Estimated cost / 1M tokens

Cheapest endpoint ranking

Rank providers by
Cheapest gpt-oss-20b Inference Providers
RankProvider / routeEstimated cost / 1M tokensSpeed measurementsEndpoint details
#1Provider / route
Estimated cost / 1M tokens$0.063333 / 1M tokensSpeed measurements
500-token response
3.48 s
Response starts
0.28 s
Output speed
156.0 tok/s
Endpoint details
Input price
$0.03 / 1M tokens
Output price
$0.13 / 1M tokens
Uptime (30 min)
99.63%
Context
131.1K
#2Provider / route
DekaLLMdekallm/bf16
Estimated cost / 1M tokens$0.066 / 1M tokensSpeed measurements
500-token response
30.32 s
Response starts
0.91 s
Output speed
17.0 tok/s
Endpoint details
Input price
$0.029 / 1M tokens
Output price
$0.14 / 1M tokens
Uptime (30 min)
96.95%
Context
131.1K
#3Provider / route
DeepInfradeepinfra/bf16
Estimated cost / 1M tokens$0.066667 / 1M tokensSpeed measurements
500-token response
5.43 s
Response starts
0.76 s
Output speed
107.0 tok/s
Endpoint details
Input price
$0.03 / 1M tokens
Output price
$0.14 / 1M tokens
Uptime (30 min)
100.00%
Context
131.1K
#4Provider / route
NovitaAInovita/fp4
Estimated cost / 1M tokens$0.076667 / 1M tokensSpeed measurements
500-token response
19.09 s
Response starts
0.57 s
Output speed
27.0 tok/s
Endpoint details
Input price
$0.04 / 1M tokens
Output price
$0.15 / 1M tokens
Uptime (30 min)
100.00%
Context
131.1K
#5Provider / route
Phalaphala
Estimated cost / 1M tokens$0.076667 / 1M tokensSpeed measurements
500-token response
9.55 s
Response starts
0.62 s
Output speed
56.0 tok/s
Endpoint details
Input price
$0.04 / 1M tokens
Output price
$0.15 / 1M tokens
Uptime (30 min)
100.00%
Context
131.1K
#6Provider / route
SiliconFlowsiliconflow/fp8
Estimated cost / 1M tokens$0.086667 / 1M tokensSpeed measurements
500-token response
46.81 s
Response starts
1.35 s
Output speed
11.0 tok/s
Endpoint details
Input price
$0.04 / 1M tokens
Output price
$0.18 / 1M tokens
Uptime (30 min)
100.00%
Context
131.1K
#7Provider / route
Parasailparasail/fp4
Estimated cost / 1M tokens$0.093333 / 1M tokensSpeed measurements
500-token response
5.58 s
Response starts
0.47 s
Output speed
98.0 tok/s
Endpoint details
Input price
$0.04 / 1M tokens
Output price
$0.2 / 1M tokens
Uptime (30 min)
96.17%
Context
131.1K
#8Provider / route
Amazon Bedrockamazon-bedrock/eu-west-1
Estimated cost / 1M tokens$0.096667 / 1M tokensSpeed measurements
500-token response
Insufficient recent data
Response starts
Insufficient recent data
Output speed
Insufficient recent data
Endpoint details
Input price
$0.07 / 1M tokens
Output price
$0.15 / 1M tokens
Uptime (30 min)
Insufficient recent data
Context
131.1K
#9Provider / route
Amazon Bedrockamazon-bedrock
Estimated cost / 1M tokens$0.096667 / 1M tokensSpeed measurements
500-token response
16.02 s
Response starts
7.69 s
Output speed
60.0 tok/s
Endpoint details
Input price
$0.07 / 1M tokens
Output price
$0.15 / 1M tokens
Uptime (30 min)
97.07%
Context
131.1K
#10Provider / route
Togethertogether
Estimated cost / 1M tokens$0.1 / 1M tokensSpeed measurements
500-token response
5.05 s
Response starts
0.27 s
Output speed
104.5 tok/s
Endpoint details
Input price
$0.05 / 1M tokens
Output price
$0.2 / 1M tokens
Uptime (30 min)
Insufficient recent data
Context
131.1K
#11Provider / route
Google Vertexgoogle-vertex/us-central1
Estimated cost / 1M tokens$0.13 / 1M tokensSpeed measurements
500-token response
6.78 s
Response starts
3.31 s
Output speed
144.0 tok/s
Endpoint details
Input price
$0.07 / 1M tokens
Output price
$0.25 / 1M tokens
Uptime (30 min)
61.56%
Context
131.1K
#12Provider / route
Fireworksfireworks
Estimated cost / 1M tokens$0.146667 / 1M tokensSpeed measurements
500-token response
4.30 s
Response starts
0.33 s
Output speed
126.0 tok/s
Endpoint details
Input price
$0.07 / 1M tokens
Output price
$0.3 / 1M tokens
Uptime (30 min)
99.06%
Context
131.1K
#13Provider / route
Groqgroq
Estimated cost / 1M tokens$0.15 / 1M tokensSpeed measurements
500-token response
1.21 s
Response starts
0.30 s
Output speed
547.0 tok/s
Endpoint details
Input price
$0.075 / 1M tokens
Output price
$0.3 / 1M tokens
Uptime (30 min)
99.91%
Context
131.1K

Endpoint data fetched .

gpt-oss-20b endpoint guide

How to interpret the Cheapest gpt-oss-20b Inference Providers

13 of 13 gpt-oss-20b endpoints currently have the published data required for this ranking. Weights & Biases leads at $0.063333 / 1M tokens via wandb/fp4. The table keeps unranked routes visible so missing measurements do not look like missing provider availability.

How estimated token cost is ranked

The cheapest ranking uses a 1,000-input/500-output-token ratio and reports the blended result per 1 million total tokens. Because providers can price input and output differently, a workload with a different token ratio may produce a different order.

Compare Weights & Biases with the next option

DekaLLM currently ranks second at $0.066 / 1M tokens. Compare that gap with input and output price, response-start time, output speed, 30-minute uptime, context length, and the exact route before deciding whether first place is meaningful for your workload.

Use the route, not only the provider name

gpt-oss-20b has 13 published provider options, and performance data is matched to each exact OpenRouter routing tag. Different quantization, context, regional deployment, or provider configuration can change price and behavior even when the underlying model name is identical.