OPENAI · SERVERLESS AND DEDICATED · LIVE
gpt-oss-120b
OpenAI's open-weight 117B mixture of experts, the catalog's default for production chat, streamed byte for byte.
131K
TOKEN CONTEXT
$0.0225
INPUT, PER MILLION
$0.1275
OUTPUT, PER MILLION
Streamed
BYTE FOR BYTE, NO BUFFER
Overview
gpt-oss-120b is OpenAI's larger open-weight model, a mixture of experts with 117B total parameters and 5.1B active per token, released under Apache 2.0. The sparse activation is what makes the economics work: flagship-adjacent quality at a per-token cost closer to a small model, and output priced well under the median listing for its class.
It is the model most teams standardize on for production chat, and the same properties carry to synthetic data generation and agent loops that burn tokens all day. The 131,072-token window covers long conversations and tool transcripts without pruning, and the weights make a clean fine-tune base for dedicated deployments.
Serving on Nyx
Every response streams byte for byte from the first token the engine produces, with nothing buffering chunks between the engine and your socket. When demand exceeds declared capacity the API answers 429 immediately with a Retry-After header rather than parking your request in a queue.
The prefix cache persists across turns, cutting 31 percent of prefill work on agent traces, which matters at this model's request volumes. Dedicated pods pin capacity with guaranteed throughput at a flat hourly price.
OUTPUT PRICE · $/M · LOWER IS BETTER
Independent harness · run weekly · streamed
# OpenAI-compatible: swap the base URL
curl https://api.nyxprovider.com/v1/chat/completions \
-H "Authorization: Bearer $NYX_API_KEY" \
-d '{
"model": "gpt-oss-120b",
"stream": true,
"messages": [{"role": "user",
"content": "Draft a support reply for this delayed order."}]
}'