OpenAI · OpenAI-compatible
gpt-oss-20b API
Fast, low-cost general model. $0.017 in, $0.085 out per 1M tokens.
| Model id | gpt-oss-20b |
| Input | $0.017 per 1M tokens |
| Output | $0.085 per 1M tokens |
| Context | 131,072 tokens |
| Input types | Text |
| Tool calling | Yes — tool_choice auto, required |
| JSON output | Yes — json_object and json_schema |
| Data | Served only on dedicated GPUs we operate. Not stored, not used for training. If no GPU is online the request fails (503), never rerouted. |
Lower on input and output than all 10 provider endpoints listing it there; the lowest, Darkbloom, is $0.018 / $0.09 per 1M tokens on OpenRouter. Checked 2026-10-31 22:40 UTC · source
Compare: vs DeepInfra
Good for
- High-volume work: classification, extraction, routing
- Tool calling and JSON output
- Chat and summaries over long inputs
Call it
curl https://synap.linkrra.com/v1/chat/completions \
-H "Authorization: Bearer $SYNAP_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "gpt-oss-20b", "messages": [{"role": "user", "content": "Hello"}]}'from openai import OpenAI
client = OpenAI(base_url="https://synap.linkrra.com/v1", api_key="sk-pool-...")
r = client.chat.completions.create(
model="gpt-oss-20b",
messages=[{"role": "user", "content": "Hello"}],
)
print(r.choices[0].message.content)Any OpenAI SDK works: change base_url to https://synap.linkrra.com/v1. Tool setup: guides.
Want to try before paying? The free tier covers qwen3-coder-30b.