GPT OSS 120B is a large-scale open-source language model designed for high-capacity reasoning, coding, and general-purpose tasks. It balances state-of-the-art performance typical of 100+ billion parameter models with relative cost efficiency, enabling broad accessibility for researchers and developers. GPT OSS 120B excels across text generation, multi-step logical reasoning, and multilingual understanding, supporting both general and specialized applications.
Key Capabilities
vs GPT-4o Mini: GPT OSS 120B offers a much larger parameter count and excels in high-capacity reasoning and code generation, while GPT-4o Mini is smaller and more cost-efficient, with built-in multimodal support for both text and images.
vs GLM-4.5: Although GLM-4.5 has more total and active parameters and leads in advanced tool integration and agentic task performance, GPT OSS 120B remains competitive with strong reasoning benchmarks and greater efficiency on smaller hardware.
Measured on the AI/ML API gateway across live traffic. Figures are medians unless stated. Last updated 2026-10-12.
import requests r = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={"Authorization": "Bearer " + AIMLAPI_KEY}, json={ "model": "openai/gpt-oss-120b", "messages": [ { "role": "user", "content": "Hello!" } ] }, ) print(r.json())
const r = await fetch("https://api.aimlapi.com/v1/chat/completions", { method: "POST", headers: { Authorization: `Bearer ${process.env.AIMLAPI_KEY}`, "Content-Type": "application/json", }, body: JSON.stringify({ "model": "openai/gpt-oss-120b", "messages": [ { "role": "user", "content": "Hello!" } ] }), }); console.log(await r.json());
curl -X POST https://api.aimlapi.com/v1/chat/completions \ -H "Authorization: Bearer $AIMLAPI_KEY" \ -H "Content-Type: application/json" \ -d '{"model":"openai/gpt-oss-120b","messages":[{"role":"user","content":"Hello!"}]}'
OpenAI-compatible — swap the base URL and it works with your existing SDK.
| Type | Price |
|---|---|
| Input | |
| Output | |
| Cached input |
| Benchmark | Score | What it measures | Source | Retrieved |
|---|---|---|---|---|
| Humanity's Last Exam | 14.9% | Expert-level questions across many domains | Source | July 8, 2026 |
| GPQA Diamond | 80.1% | Google-proof graduate science questions (hardest subset) | Source | July 8, 2026 |
| AIME 2025 | 92.5% | Competition mathematics (AIME), 2025 | Source | July 8, 2026 |
| Intelligence | 12.3 | Composite score across standardised reasoning, knowledge and problem-solving evaluations, measured independently by Artificial Analysis | Source | September 12, 2026 |
| Coding | 30.4 | Composite score across standardised coding evaluations, measured independently by Artificial Analysis | Source | September 12, 2026 |
| Math | 93.4 | Composite score across standardised mathematics evaluations, measured independently by Artificial Analysis | Source | September 12, 2026 |
| Model | Input | Output | Context | Best for |
|---|---|---|---|---|
GPT OSS 120B This page | Reasoning + agents | |||
| Reasoning + agents | ||||
| Balanced coding + agents | ||||
| Long-context, multimodal & agentic workflows | ||||
| Reasoning + agents |
GPT OSS 120B has a 131,000 tokens context window and can return up to 131,072 tokens.
GPT OSS 120B takes text as input and returns text.
Use openai/gpt-oss-120b as the model id. Requests go to https://api.aimlapi.com/v1/chat/completions.
GPT OSS 120B became available on September 9, 2025.
GPT OSS 120B is priced at input $0.48139 / 1M tokens, output $1.30663 / 1M tokens, cached input $0.48139 / 1M tokens.
Yes, GPT OSS 120B can stream responses as they are generated.
GPT OSS 120B was built by OpenAI.
It is accessed via the endpoint https://api.aimlapi.com/v1/chat/completions, not directly through OpenAI's own API.
