gpt-oss-120b
openai/gpt-oss-120bby OpenAI·Text·131K context·Released 2025-08-05·Apache 2.0
Overview
The larger sibling of gpt-oss-20b: a 117B-parameter open-weight MoE (5.1B active) under Apache 2.0. Noticeably stronger reasoning, coding and instruction-following while staying inexpensive to serve thanks to the sparse architecture. Hosted by a dozen providers for competitive live pricing and failover. A common open default when the 20B model isn't quite strong enough but you still want to avoid first-party frontier prices.
Benchmarks
Standardized scores (0–100), normalized across the catalog.
Intelligence58
Coding74
Instruction following90
Providers
LLMRouteHub routes to the cheapest healthy provider by default. Showing the top 5 of 12 providers by routed volume.
| Provider | Input | Output | Throughput | Uptime |
|---|---|---|---|---|
| DeepInfra | $0.10 | $0.50 | 88 tok/s | 99.8% |
| Together | $0.12 | $0.61 | 82 tok/s | 99.6% |
| Fireworks | $0.15 | $0.73 | 76 tok/s | 99.5% |
| Novita | $0.15 | $0.77 | 70 tok/s | 99.4% |
| Groq | $0.18 | $0.89 | 63 tok/s | 99.2% |
Use this model
curl https://llmroutehub.com/api/v1/chat/completions \
-H "Authorization: Bearer $LLMROUTEHUB_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "openai/gpt-oss-120b",
"messages": [{ "role": "user", "content": "Hello" }]
}'