GPT OSS 120B

gpt-oss-120b by OpenAI Released August 6, 2025
ReasoningFunction callingStructured outputPrompt cachingWeb searchComputer useImage inputFile inputAudio inputVideo inputEU hostingFree endpointUp to % off
Input from
$0.03 / 1M
Output from
$0.17 / 1M
Cached input from
$0.0045 / 1M
Context window
131K
Endpoints
17
from 14 providers
Regions
Global, Europe, United States

About GPT OSS 120B

GPT OSS 120B is a state-of-the-art, reasoning model with chain-of-thought and adjustable reasoning effort levels built and trained by OpenAI. It is OpenAI's flagship open-weight model that features a 128K token context window. The model is built for high-quality reasoning tasks.

GPT OSS 120B is a chat model by OpenAI, available on Eden AI through 17 endpoints from 14 providers, from $0.03 per million input tokens and $0.17 per million output tokens, with a context window of up to 131K tokens.

Providers and prices

17 endpoints from 14 providers, cheapest first, in USD per million tokens. Every one is called with the same Eden AI key: pin one with its endpoint ID, or let Eden AI route for you. .

Endpoint
Region
Context
Input / 1M
Output / 1M
Cached / 1M
Supports
FlexAIOfficial−%flexai/gpt-oss-120b
Global
131K
$0.03$0.03
$0.17
$0.0045
ReasoningToolsJSONCacheVisionWeb
DeepInfra (United States)Official−%deepinfra/openai/gpt-oss-120b
United States
131K
$0.037$0.037
$0.17
$
ReasoningToolsJSONCacheVisionWeb
OVHcloud (Europe)Official−%ovhcloud/gpt-oss-120b
Europe
131K
$0.09$0.09
$0.47
$
ReasoningToolsJSONCacheVisionWeb
Google CloudOfficial−%vertex/openai/gpt-oss-120b-maas
Global
131K
$0.09$0.09
$0.36
$
ReasoningToolsJSONCacheVisionWeb
Databricks (Europe)Official−%databricks/databricks-gpt-oss-120b@eu
Europe
131K
$0.15$0.15
$0.6
$0.015
ReasoningToolsJSONCacheVisionWeb
DatabricksOfficial−%databricks/databricks-gpt-oss-120b
Global
131K
$0.15$0.15
$0.6
$0.015
ReasoningToolsJSONCacheVisionWeb
Fireworks AI (United States)Official−%fireworks_ai/accounts/fireworks/models/gpt-oss-120b
United States
131K
$0.15$0.15
$0.6
$0.015
ReasoningToolsJSONCacheVisionWeb
Amazon Web Services (Europe)Official−%amazon/openai.gpt-oss-120b-1:0
Europe
128K
$0.15$0.15
$0.6
$
ReasoningToolsJSONCacheVisionWeb
Amazon Web Services (United States)Official−%amazon/openai.gpt-oss-120b-1:0@us
United States
128K
$0.15$0.15
$0.6
$
ReasoningToolsJSONCacheVisionWeb
Fireworks AI (United States)Official−%fireworks_ai/gpt-oss-120b
United States
131K
$0.15$0.15
$0.6
$0.014
ReasoningToolsJSONCacheVisionWeb
Groq (United States)Official−%groq/openai/gpt-oss-120b
United States
131K
$0.15$0.15
$0.6
$0.075
ReasoningToolsJSONCacheVisionWeb
NebiusOfficial−%nebius/openai/gpt-oss-120b
Global
131K
$0.15$0.15
$0.6
$0.15
ReasoningToolsJSONCacheVisionWeb
Together AIOfficial−%together_ai/openai/gpt-oss-120b
Global
131K
$0.15$0.15
$0.6
$
ReasoningToolsJSONCacheVisionWeb
IONOS (Europe)Official−%ionos/openai/gpt-oss-120b
Europe
$0.1681$0.1681
$0.7284
$
ReasoningToolsJSONCacheVisionWeb
Scaleway (Europe)Official−%scaleway/gpt-oss-120b
Europe
128K
$0.1681$0.1681
$0.6724
$
ReasoningToolsJSONCacheVisionWeb
Cloudflare (CF/OPENAI/GPT-OSS-120B)Official−%cloudflare/@cf/openai/gpt-oss-120b
CF/OPENAI/GPT-OSS-120B
128K
$0.35$0.35
$0.75
$
ReasoningToolsJSONCacheVisionWeb
Cerebras (United States)Official−%cerebras/gpt-oss-120b
United States
131K
$0.35$0.35
$0.75
$0.35
ReasoningToolsJSONCacheVisionWeb

Performance

Measured on real Eden AI traffic over the last 30 days. Endpoints with too little traffic are not shown.

Fastest first token
0.22 s
Scaleway (Europe)
Lowest latency
0.39 s
Cerebras (United States)
Highest throughput
395.52 tok/s
Groq (United States)
Best availability
100%
Databricks (Europe)
Endpoint
First token
Latency p50
Latency p95
Throughput
Availability
Cache hits
Tool errors
JSON errors
FlexAI
0.99 s
14.4 s
33 s
109.47 tok/s
99.98%
32.66%
7.64%
0%
DeepInfra (United States)
0.55 s
10.93 s
24.74 s
41.1 tok/s
100%
0.16%
0%
0%
OVHcloud (Europe)
0.73 s
2.17 s
52.62 s
41.22 tok/s
99.5%
0%
0%
0%
Google Cloud
3.32 s
5.87 s
183.67 s
11.22 tok/s
95%
0%
50%
46.67%
Databricks (Europe)
s
3 s
13.76 s
129.22 tok/s
100%
0.18%
4.76%
100%
Databricks
1.3 s
4.61 s
14.69 s
144.92 tok/s
100%
0.2%
3.57%
100%
Fireworks AI (United States)
s
s
s
tok/s
%
%
%
%
Amazon Web Services (Europe)
0.42 s
3.67 s
8.13 s
160.7 tok/s
99.93%
0%
0%
0.15%
Amazon Web Services (United States)
s
s
s
tok/s
%
%
%
%
Fireworks AI (United States)
s
s
s
tok/s
%
%
%
%
Groq (United States)
0.92 s
2.49 s
5.05 s
395.52 tok/s
99.82%
16.61%
17.16%
1.12%
Nebius
0.29 s
2.38 s
19.62 s
210.94 tok/s
99.97%
29.27%
5.56%
0%
Together AI
0.29 s
2.27 s
8.73 s
94.88 tok/s
82.61%
36.78%
0%
23.53%
IONOS (Europe)
0.28 s
1.46 s
4.29 s
125.9 tok/s
79.07%
0%
1.64%
46.55%
Scaleway (Europe)
0.22 s
6.59 s
37.97 s
89.91 tok/s
79.23%
0%
0%
66.32%
Cloudflare (CF/OPENAI/GPT-OSS-120B)
s
s
s
tok/s
%
%
%
%
Cerebras (United States)
0.22 s
0.39 s
1.64 s
170.8 tok/s
100%
19.96%
6.76%
0%

Estimate your monthly cost

Enter your expected volume to compare every endpoint at once.

Call GPT OSS 120B with Eden AI

Use the model ID to let Eden AI pick an endpoint, or an endpoint ID from the table above to pin one.

import requests

response = requests.post(
    "https://api.edenai.run/v3/chat/completions",
    headers={"Authorization": "Bearer YOUR_EDENAI_API_KEY"},
    json={
        "model": "",
        "messages": [{"role": "user", "content": "Hello!"}],
    },
)
print(response.json())

Questions about GPT OSS 120B

On Eden AI, prices for this model start at $0.03 per million input tokens and $0.17 per million output tokens. Prices differ by endpoint; the providers table lists each one. .

Up to 131K tokens, depending on the endpoint.

17 endpoints from 14 providers, available in: Global, Europe, United States. All of them are called with the same Eden AI API key.

Yes. At least one endpoint is hosted in Europe; pin it with its endpoint ID from the providers table to keep requests in the EU.

Over the last 30 days, the fastest first token was 0.22 s (Scaleway (Europe)) and the highest throughput was 395.52 tokens per second (Groq (United States)).