Requesty
Fireworks AI

gpt-oss-20b

GPT-OSS-20B is OpenAI's lightweight open-weight model with 21.5 billion total parameters (3.6 billion active, Mixture-of-Experts). Optimized for lower latency and local or specialized use cases including agentic tasks, chain-of-thought reasoning, and web browsing. Released under Apache 2.0 license.

🧠Reasoning⚡Caching
Try gpt-oss-20bCompareDocs

Specifications

Context window131K tokens
Max output66K tokens
API typechat
AddedMay 27, 2026
Model IDfireworks/gpt-oss-20b
Data retentionNo
Used for trainingNo
Provider location🇺🇸 US
Privacy policyFireworks Privacy Policy→

Benchmarks

Benchmarks haven't been published yet for this exact variant.

Some variants (region-specific deployments, highspeed tiers) share benchmarks with their base model — check the base model page or the Fireworks AI models overview.

Pricing

Input / 1M
$0.07
Output / 1M
$0.30
Cache write
—
Cache read / 1M
$0.04
Estimated cost
100K input + 10K output$0.0100
1M input + 100K output$0.10
10M input + 1M output$1.00

Requesty charges exactly what the upstream provider charges — no markup, no per-request fees. Prompt caching and smart routing can reduce effective cost by 30-80%.

Quickstart

Drop-in compatible with the OpenAI SDK. Change the base URL, swap in your Requesty API key, and set the model to fireworks/gpt-oss-20b.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_REQUESTY_API_KEY",
    base_url="https://router.requesty.ai/v1",
)

response = client.chat.completions.create(
    model="fireworks/gpt-oss-20b",
    messages=[
        {"role": "user", "content": "Explain quantum computing in one paragraph."},
    ],
)

print(response.choices[0].message.content)

Other Fireworks AI models

gpt-oss-120b
131K ctx$0.15/1M
deepseek-v4-flash
1M ctx$0.14/1M
qwen3.6-plus
131K ctx$0.50/1M
minimax-m2.7
197K ctx$0.30/1M
glm-5.1
202K ctx$1.40/1M
deepseek-v4-pro
1M ctx$1.74/1M

Frequently asked questions

How much does gpt-oss-20b cost?
gpt-oss-20b is priced at $0.07 per million input tokens and $0.30 per million output tokens when accessed via Requesty. Prompt caching is supported, which can cut effective input cost by up to 90% on repeated context. Requesty charges exactly what the upstream provider charges — we don't add markup.
What is the context window of gpt-oss-20b?
gpt-oss-20b has a context window of 131K tokens, with a maximum output of 66K tokens per response. That's roughly 175 words of input you can fit in a single prompt.
What can gpt-oss-20b do?
gpt-oss-20b supports extended reasoning, prompt caching. You can call it through any OpenAI-compatible client by pointing base_url to Requesty.
How do I use gpt-oss-20b with the OpenAI SDK?
Install the OpenAI SDK, set base_url to "https://router.requesty.ai/v1", set your API key to your Requesty key, and set the model to "fireworks/gpt-oss-20b". The Quickstart above shows Python, JavaScript and cURL snippets.

Access gpt-oss-20b through Requesty

One API key, 400+ models, OpenAI-compatible. No markup on provider prices, automatic failover, and smart caching built-in.

Get started freeAll Fireworks AI models