# OpenAI GPT 120B

> OpenAI GPT 120B is an open-weight language model from OpenAI, released on 5 Aug 2025. It accepts text and generates text, with a 128K-token context window and up to 128K output tokens per response. As of 4 Oct 2026, the lowest listed API price is $0.04 per million input tokens and $0.17 per million output tokens (DeepInfra), across 15 providers we track.

Source page: https://www.aimodel.directory/models/gpt-oss-120b
Last verified: 4 Oct 2026

## Specifications

- **Developer:** OpenAI
- **Release date:** 5 Aug 2025 (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/openai/gpt-oss-120b.toml)
- **Status:** Generally available (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/openai/gpt-oss-120b.toml)
- **Weights:** Open weights (source: Hugging Face, https://huggingface.co/openai/gpt-oss-120b)
- **License:** apache-2.0 (source: Hugging Face, https://huggingface.co/openai/gpt-oss-120b)
- **Parameters:** 117B (source: Hugging Face, https://huggingface.co/openai/gpt-oss-120b)
- **Context window:** 128K (128,072 tokens) (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/openai/gpt-oss-120b.toml)
- **Max output:** 128K tokens (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/openai/gpt-oss-120b.toml)
- **Knowledge cutoff:** Aug 2025 (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/openai/gpt-oss-120b.toml)
- **Input:** Text (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/openai/gpt-oss-120b.toml)
- **Output:** Text (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/openai/gpt-oss-120b.toml)
- **Reasoning mode:** Yes (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/openai/gpt-oss-120b.toml)
- **Tool calling:** Yes (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/openai/gpt-oss-120b.toml)
- **Structured output:** Yes (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/openai/gpt-oss-120b.toml)
- **Hugging Face:** openai/gpt-oss-120b (source: Hugging Face, https://huggingface.co/openai/gpt-oss-120b)

## API pricing (USD per 1M tokens)

| Provider | Input | Output | Cached input | Context |
|---|---|---|---|---|
| DeepInfra | $0.04 | $0.17 | – | 131K |
| OpenRouter | $0.04 | $0.17 | – | 131K |
| SiliconFlow | $0.05 | $0.45 | – | 131K |
| Novita AI | $0.05 | $0.25 | – | 131K |
| Google Vertex AI | $0.09 | $0.36 | – | 131K |
| Baseten | $0.10 | $0.50 | – | 128K |
| Vercel AI Gateway | $0.10 | $0.50 | $0.10 | 131K |
| Amazon Bedrock | $0.15 | $0.60 | – | 131K |
| Amazon Bedrock | $0.15 | $0.60 | – | 131K |
| Groq | $0.15 | $0.60 | $0.07 | 131K |
| Nebius AI Studio | $0.15 | $0.60 | $0.01 | 131K |
| Fireworks AI | $0.15 | $0.60 | $0.01 | 131K |
| Together AI | $0.15 | $0.60 | – | 131K |
| Amazon Bedrock | $0.18 | $0.72 | – | 128K |
| Hugging Face Inference Providers | $0.25 | $0.69 | – | 131K |
| Cloudflare Workers AI | $0.35 | $0.75 | – | 128K |
| Cerebras | $0.35 | $0.75 | – | 131K |

## FAQ

### What is the context window of OpenAI GPT 120B?

OpenAI GPT 120B has a context window of 128,072 tokens (128K) and can generate up to 128,072 tokens in a single response, according to models.dev.

### How much does the OpenAI GPT 120B API cost?

OpenAI GPT 120B is listed from $0.04 per million input tokens and $0.17 per million output tokens (DeepInfra). Prices last verified 4 Oct 2026.

### Is OpenAI GPT 120B open source?

OpenAI GPT 120B is an open-weight model: its weights are published on Hugging Face as openai/gpt-oss-120b under the apache-2.0 license. Check the license terms before commercial use.

### When was OpenAI GPT 120B released?

OpenAI released OpenAI GPT 120B on 5 Aug 2025, according to models.dev.

### Which providers offer OpenAI GPT 120B?

We track OpenAI GPT 120B on 15 providers: DeepInfra, OpenRouter, SiliconFlow, Novita AI, Google Vertex AI, Baseten, Vercel AI Gateway, Amazon Bedrock, Groq, Nebius AI Studio, Fireworks AI, Together AI, Hugging Face Inference Providers, Cloudflare Workers AI, Cerebras.

---
Data licensed CC BY 4.0. Cite as: AI Model Directory, "OpenAI GPT 120B", https://www.aimodel.directory/models/gpt-oss-120b