gpt-oss-20b
Coming soonThe small open-weights gpt-oss model, light and cheap to run.
Context window—
Input$0.0690 IQDper 1M tokens
Output$0.225320 IQDper 1M tokens
Input modalitiesText→ Text
Providers12
Overview
| Model ID | openai/gpt-oss-20b |
|---|---|
| Series | OpenAI |
| Context window | — |
| Max output | — |
| Input modalities | Text |
| Output modalities | Text |
| Providers | 12 |
Price on OPTES
Our price, per provider, in US dollars and Iraqi dinars (1 USD = 1,380 IQD, rounded up to 10 IQD).
| Provider | Input | Output | Note |
|---|---|---|---|
| Novita AI | $0.06 · 90 IQD | $0.225 · 320 IQD | Lowest |
Providers
Our price, per provider, in US dollars and Iraqi dinars (1 USD = 1,380 IQD, rounded up to 10 IQD).
| Provider | Input | Output | Note |
|---|---|---|---|
| CoreWeave | $0.045 · 70 IQD | $0.195 · 270 IQD | Approximatereported through OpenRouter's listing, 2026-10-01 |
| AkashML | $0.045 · 70 IQD | $0.195 · 270 IQD | ApproximateAkashML docs say prices may lag the live API |
| DeepInfra | $0.045 · 70 IQD | $0.21 · 290 IQD | Approximatereported through OpenRouter's listing, 2026-10-01 |
| Novita AI | $0.06 · 90 IQD | $0.225 · 320 IQD | Lowest |
| SiliconFlow | $0.06 · 90 IQD | $0.27 · 380 IQD | |
| Parasail | $0.06 · 90 IQD | $0.30 · 420 IQD | |
| OVHcloud AI Endpoints | $0.075 · 110 IQD | $0.27 · 380 IQD | |
| Hugging Face Inference Providers | $0.075 · 110 IQD | $0.30 · 420 IQD | via Nscale |
| Nscale | $0.075 · 110 IQD | $0.30 · 420 IQD | Approximatereported through the Hugging Face router, 2026-10-01 |
| io.net | $0.0945 · 140 IQD | $0.294 · 410 IQD | |
| Groq | $0.1125 · 160 IQD | $0.45 · 630 IQD | ~1000 tokens/s |
| Cloudflare Workers AI | $0.30 · 420 IQD | $0.45 · 630 IQD |
Call it through the OPTES API
This model is coming soon to OPTES. This is how the call will look; your OPTES key will work for it.
curl https://api.optes3.com/v1/chat/completions \ -H "Authorization: Bearer $OPTES_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "openai/gpt-oss-20b", "messages": [{"role": "user", "content": "Suggest three names for a café in Baghdad."}] }'import osfrom openai import OpenAIclient = OpenAI(base_url="https://api.optes3.com/v1", api_key=os.environ["OPTES_API_KEY"])resp = client.chat.completions.create( model="openai/gpt-oss-20b", messages=[{"role": "user", "content": "Suggest three names for a café in Baghdad."}],)print(resp.choices[0].message.content)import OpenAI from "openai";const client = new OpenAI({ baseURL: "https://api.optes3.com/v1", apiKey: process.env.OPTES_API_KEY });const resp = await client.chat.completions.create({ model: "openai/gpt-oss-20b", messages: [{ role: "user", content: "Suggest three names for a café in Baghdad." }],});console.log(resp.choices[0].message.content);Coming soon to OPTES Coming soon
Get your OPTES API key now. The same key will call this model when it opens.