Free tool
Context Window Comparison: How Much Fits?
GPT-6 Astra accepts 1.05M tokens, about 1,575 pages of text.
Last verified
- GPT-6 Astra 1,050K
- GPT-6.1 Sol 1,050K
- GPT-5.6 Sol 1,050K
- GPT-5.5 Pro 1,050K
- GPT-5.5 1,050K
- GPT-6 Sol 1,050K
- GPT-5.4 1,050K
- GPT-5.6 Terra 1,050K
- GPT-5.4 Pro 1,050K
- GPT-5.6 Luna 1,050K
- GPT-6 Luna 1,050K
- Gemini 3.8 Flash 1,049K
- Gemini 3.7 Flash 1,049K
- Kimi K3 1,049K
- Gemini 3.1 Pro Preview 1,049K
| Model | Context | ≈ Pages | ≈ Novels | Max output |
|---|---|---|---|---|
| GPT-6 Astra | 1.05M | 1,575 | 8.8 | 128K |
| GPT-6.1 Sol | 1.05M | 1,575 | 8.8 | 128K |
| GPT-5.6 Sol | 1.05M | 1,575 | 8.8 | 128K |
| GPT-5.5 Pro | 1.05M | 1,575 | 8.8 | 128K |
| GPT-5.5 | 1.05M | 1,575 | 8.8 | 128K |
| GPT-6 Sol | 1.05M | 1,575 | 8.8 | 128K |
| GPT-5.4 | 1.05M | 1,575 | 8.8 | 128K |
| GPT-5.6 Terra | 1.05M | 1,575 | 8.8 | 128K |
| GPT-5.4 Pro | 1.05M | 1,575 | 8.8 | 128K |
| GPT-5.6 Luna | 1.05M | 1,575 | 8.8 | 128K |
| GPT-6 Luna | 1.05M | 1,575 | 8.8 | 128K |
| Gemini 3.8 Flash | 1.05M | 1,573 | 8.7 | 66K |
| Gemini 3.7 Flash | 1.05M | 1,573 | 8.7 | 66K |
| Kimi K3 | 1.05M | 1,573 | 8.7 | 1.05M |
| Gemini 3.1 Pro Preview | 1.05M | 1,573 | 8.7 | 66K |
| Muse Spark 1.3 | 1.05M | 1,573 | 8.7 | 131K |
| Gemini 3.5 Flash | 1.05M | 1,573 | 8.7 | 66K |
| Gemini 3.6 Flash | 1.05M | 1,573 | 8.7 | 66K |
| Gemini 3 Flash Preview | 1.05M | 1,573 | 8.7 | 66K |
| Muse Spark 1.2 | 1.05M | 1,573 | 8.7 | 131K |
| MiMo-V2.6-Pro | 1.05M | 1,573 | 8.7 | 131K |
| Muse Spark 1.1 | 1.05M | 1,573 | 8.7 | 131K |
| MiMo-V2.6-Flash | 1.05M | 1,573 | 8.7 | 131K |
| Hy4 preview | 1.05M | 1,573 | 8.7 | 64K |
| MiMo-V2.5-Pro | 1.05M | 1,573 | 8.7 | 131K |
| Gemini 2.5 Pro | 1.05M | 1,573 | 8.7 | 66K |
| MiMo-V2.5 | 1.05M | 1,573 | 8.7 | 131K |
| Mistral Large 4 | 1.05M | 1,573 | 8.7 | 262K |
| MiMo-V2-Pro | 1.05M | 1,573 | 8.7 | 131K |
| Gemini 3.5 Flash Lite | 1.05M | 1,573 | 8.7 | 66K |
| Gemini 3.1 Flash Lite | 1.05M | 1,573 | 8.7 | 66K |
| Gemini 2.5 Flash | 1.05M | 1,573 | 8.7 | 66K |
| Gemini 2.5 Flash-Lite | 1.05M | 1,573 | 8.7 | 66K |
| GPT-4.1 | 1.05M | 1,571 | 8.7 | 33K |
| GPT-4.1 mini | 1.05M | 1,571 | 8.7 | 33K |
| GPT-4.1 nano | 1.05M | 1,571 | 8.7 | 33K |
| Step 5 Preview | 1.02M | 1,536 | 8.5 | 66K |
| Claude Fable 5.1 | 1M | 1,500 | 8.3 | 128K |
| Claude Opus 5.5 | 1M | 1,500 | 8.3 | 128K |
| Claude Opus 5 | 1M | 1,500 | 8.3 | 128K |
Frequently asked questions
Which AI model has the largest context window?
GPT-6 Astra accepts 1.05M tokens, about 1,575 pages of text.
Does a bigger context window mean better answers on long documents?
No. Many models lose accuracy well before their limit. Check the long-context category ranking, which measures how well models actually use long inputs.