including and which point to this snapshot (gpt-4-0125-preview) shut down on 2026-03-26 per OpenAI's deprecation page (checked 2026-10-17). The official replacement is gpt-5. The published context window changes from 128,000 tokens on gpt-4-0125-preview to 400,000 tokens on gpt-5.
retired
| Announced | 2025-09-26 |
|---|---|
| Deprecated | Not published |
| Shutdown | 2026-03-26 |
| Kind | exact |
See the replacement diff · OpenAI models
| Input price / 1M | Not published |
|---|---|
| Output price / 1M | Not published |
| Cached input / 1M | Not published |
| Context window | 128,000 · source, checked 2026-10-17 |
| Max output | 4,096 · source, checked 2026-10-17 |
| Reasoning | Not published |
| Vision | Not published |
| Tools | Not published |
including and which point to this snapshot (gpt-4-0125-preview) shut down on 2026-03-26 per OpenAI's deprecation page (checked 2026-10-17). The official replacement is gpt-5. The published context window changes from 128,000 tokens on gpt-4-0125-preview to 400,000 tokens on gpt-5.
The official replacement is gpt-5.
including and which point to this snapshot (gpt-4-0125-preview) status is retired.
2026‑03‑26 | gpt-4-0125-preview (including gpt-4-turbo-preview and gpt-4-turbo-preview-completions, which point to this snapshot) | gpt-5 or gpt-4.1*
Model ID: gpt-4-turbo-preview | context window 128000 | max output 4096 | input $10 | output $30 | Default snapshot gpt-4-0125-preview