GPT — free API
Updated:
Stable pages for current GPT models with free or promotional API routes, including GPT-5.6 Luna.
Read more
The list is built from current provider API responses. The same model may appear more than once because each endpoint has its own limits, context window, and free-tier rules. Individual model pages use stable URLs and update automatically.
7 models found
| № | Model | Power | Provider | Parameters | Context | Limit | |
|---|---|---|---|---|---|---|---|
| 1 |
GPT OSS 120Bopenai/gpt-oss-120b
|
🚀 | Groq | 120B | 131K | 14,400 req/day | Open |
| 2 |
GPT OSS 20Bopenai/gpt-oss-20b
|
⚖️ | Groq | 20B | 131K | 14,400 req/day | Open |
| 3 |
GPT OSS 120Bgpt-oss-120b
|
🚀 | Cerebras | 120B | 33K | 14,400 req/day · 30 req/min | Open |
| 4 |
GPT Oss 20bopenai/gpt-oss-20b
|
⚖️ | NVIDIA Build | 20B | ? | Free serverless inference for development; limits and model availability may change. | Open |
| 5 |
GPT Oss 120bgpt-oss-120b:free
|
🚀 | UnoRouter | 120B | 131K | No-card free models use shared capacity and per-model limits; availability can change. | Open |
| 6 |
GPT Oss 20bgpt-oss-20b:free
|
⚖️ | UnoRouter | 20B | 128K | No-card free models use shared capacity and per-model limits; availability can change. | Open |
| 7 |
GPT Oss Safeguard 20bgpt-oss-safeguard-20b:free
|
⚖️ | UnoRouter | 20B | 131K | No-card free models use shared capacity and per-model limits; availability can change. | Open |