gpt-oss-120b on Replicate API

GPT-OSS · OpenAI

ServerlessOpen Source

Last refreshed 2026-04-19. Next refresh: weekly.

Why use gpt-oss-120b on Replicate API?

Replicate API offers gpt-oss-120b with pay-as-you-go pricing at $0.18/1M input tokens. Replicate is a cloud-based platform that enables users to run machine learning models easily and efficiently.

Compare gpt-oss-120b across 13 providers to find the best fit for your use case
Input / 1M
$0.18
Output / 1M
$0.72
Cache
Not sourced
Batch
Not sourced

Setup recipe

Python + curl
Install
pip install replicate
Auth
export REPLICATE_API_TOKEN=...
Call
import replicate
output = replicate.run(
    "openai/gpt-oss-120b",
    input={"prompt": "Hello"}
Model ID
openai/gpt-oss-120b

Request example

import replicate

# reads REPLICATE_API_TOKEN from env
# openai/gpt-oss-120b format: "owner/model-name" (latest version) or "owner/model-name:version-hash"
output = replicate.run(
    "openai/gpt-oss-120b",
    input={"prompt": "Hello"}
)
# Output is a list or generator depending on the model
print("".join(output))

Gotchas

  • Use provider model ID "openai/gpt-oss-120b", not the LLMReference slug "gpt-oss-120b".
  • Replicate uses "owner/model-name" format (e.g. "meta/meta-llama-3-8b-instruct") for the latest version, or "owner/model-name:version-sha" to pin to a specific version. The REST endpoint splits owner and model-name into the path: /v1/models/{owner}/{model-name}/predictions.
  • The examples expect REPLICATE_API_TOKEN; rename it only if your application config maps the new variable.

Compare gpt-oss-120b Across Providers

ProviderInput (per 1M)Output (per 1M)
Cloudflare Workers AI$0.35$0.75
OpenRouter$0.04$0.18
Together AI$0.15$0.60
Fireworks AI$0.15$0.60
GCP Vertex AI$0.09$0.36
View all 13 providers →

Pricing

TypePrice (per 1M)
Input tokens$0.18
Output tokens$0.72

Capabilities

JSON / Tool useStructured Outputs

About gpt-oss-120b

OpenAI open-weight model with 120 billion parameters. Text-only model supporting reasoning, function calling, and structured outputs. Free for self-hosting. Released August 5, 2025.

Get Started

Model Specs

Released2025-08-05
Parameters120B
Context131k
ArchitectureDecoder Only
Knowledge cutoff2025-08

Related Models on Replicate API