GPT-OSS 120B API
128K context · Released Aug 5, 2025 · By OpenAI · Apache 2.0
GPT-OSS 120B costs $0.09 per million input tokens and $0.36 per million output tokens, against an official rate of $0.15 and $0.6 — 40% below official pricing. It is a OpenAI model under the Apache 2.0 licence, with a 128K context window, callable through an OpenAI-compatible endpoint with the model id `openai/gpt-oss-120b`. This model is listed for reference and is not currently served by an upstream provider, so requests for it return no available channel.
Not currently available. No configured upstream serves this model, so a request for it returns no available channel. The rate below is what it will cost when it is enabled.
Input
$0.090/M
$0.15/M
Output
$0.36/M
$0.60/M
Savings
40%
off official
- No markup
- OpenAI SDK compatible
- No minimum spend
- Apache 2.0
OpenAI's open-weight release under Apache 2.0. Familiar behaviour and prompt ergonomics if you're migrating off the closed OpenAI models.
Copy this and go
The model id is already filled in. Nothing in this snippet needs editing except your key.
from openai import OpenAI
client = OpenAI(
api_key="YOUR_API_KEY",
base_url="https://router.xark.io/api/v1" # ← the only line that changes
)
response = client.chat.completions.create(
model="openai/gpt-oss-120b",
messages=[{"role": "user", "content": "Hello"}]
)
GPT-OSS 120B pricing across providers
Rates as published by each provider. We update these when they change; if you spot one that's stale, tell us and we'll fix it.
| Provider | Input | Output |
|---|---|---|
| AI Token Router | $0.090/M | $0.36/M |
| Official (OpenAI) | $0.15/M | $0.60/M |
| OpenRouter | $0.14/M | $0.58/M |
| DeepInfra | $0.10/M | $0.45/M |
| Together AI | $0.13/M | $0.55/M |
Cached input on GPT-OSS 120B is billed at $0.018/M — roughly 80% below the standard input rate. For agent loops with a large fixed system prompt this is usually the number that matters most.
What it's good at
- Drop-in migration from closed OpenAI models
- Reasoning-effort-controlled workloads
- On-prem-compatible architectures
Other text models
GPT-OSS 120B questions
Every figure below is read from the same catalogue entry as the table above, so the two cannot disagree.
What does GPT-OSS 120B cost on AI Token Router, and what is the official rate?
Input is $0.090/M here against $0.15/M official, and output is $0.36/M against $0.60/M official — $0.060/M off every million input tokens and $0.24/M off every million output tokens. The headline 40% blends the two, weighted three to one towards output, because real workloads write far more than they read and a flat average of the two percentages would overstate what you actually save. Both figures are printed side by side on this page: the official rate is the only benchmark that makes ours mean anything.
Can I call GPT-OSS 120B right now?
No. GPT-OSS 120B is catalogued but not currently servable — no configured upstream provider carries it, so a request naming `openai/gpt-oss-120b` returns `no available channel` instead of a result. The rate on this page is what it will cost once an upstream serves it, not something you can spend today. The text models you can call today are Kimi K2.6, Kimi K3, GLM-5.2, DeepSeek V4 Pro and DeepSeek V4 Flash. 7 of the 24 catalogued models are callable, and every page says which it is.
Does GPT-OSS 120B support prompt caching, and what does a cached token cost?
Yes. A repeated prefix — a fixed system prompt, a tool schema, a document you keep querying — is served from the model's KV cache and billed at $0.018/M instead of the standard $0.090/M, which is 80% below the input rate. OpenAI's own cached rate is $0.030/M, so the discount applies on top of a rate that is already below official. Nothing has to be set on the request; the prefix is matched automatically. On an agent loop, where most of every request is unchanged from the last one, this is usually the largest single line on the bill.
How do I call GPT-OSS 120B from my code?
Point an OpenAI SDK at `https://router.xark.io/api/v1` and pass `model="openai/gpt-oss-120b"` — the chat completions endpoint is `https://router.xark.io/api/v1/chat/completions`. Request shape, streaming frames, the usage block and the error envelope all match what your OpenAI client already handles, so the base URL and the model id are the only two lines that change. The bare name `gpt-oss-120b`, without the provider prefix, resolves to the same model. An integration written before the ids were namespaced therefore keeps working.
How large is GPT-OSS 120B's context window?
128K — 131,072 tokens in a single request, shared between everything you send and everything the model generates back. For comparison, the widest context in this catalogue is Llama 4 Maverick at 1M. Context is billed like any other input: filling the window costs the input rate on every token in it, every turn, which is why the cached-input rate matters more than the window size on a workload that resends the same prefix.
What licence is GPT-OSS 120B under, and can I use it commercially?
GPT-OSS 120B is published by OpenAI under the Apache 2.0 licence. Apache 2.0 is a standard permissive licence: commercial use, modification and redistribution are allowed, subject to keeping the notices and stating what you changed. It also carries an explicit patent grant from the contributors, which terminates if you bring a patent claim over the work — that grant is usually the reason a legal team prefers it to MIT. No user threshold and no field-of-use restriction apply.
How does GPT-OSS 120B pricing here compare with other providers carrying it?
Of the 3 other providers we track carrying GPT-OSS 120B, the lowest is DeepInfra at $0.10/M input and $0.45/M output, against our $0.090/M input and $0.36/M output. That is where the rates stood when we last reconciled them, and it can change without notice — we are not claiming to be the cheapest source for this model, only that our rate sits below OpenAI's official $0.15/M. If you find one of these figures stale, tell us and we will correct it.
What is GPT-OSS 120B actually good at?
OpenAI's open-weight release under Apache 2.0. Familiar behaviour and prompt ergonomics if you're migrating off the closed OpenAI models. In practice that points at drop-in migration from closed OpenAI models, reasoning-effort-controlled workloads and on-prem-compatible architectures. The specifics that actually decide a fit — the rate, the licence and what it will and will not serve today — are all on this page rather than behind a sales conversation.