LLMs with the longest context windows
Every model we track, sorted by the maximum number of tokens it accepts in one request.
Context windows grew from a few thousand tokens in 2022 to a million or more for many current models. The limit covers your prompt, any documents or conversation history you include, and often the response.
The advertised limit is a ceiling, not a guarantee of quality across the whole window. Accuracy on facts deep inside long prompts varies, and long prompts cost more and take longer to process.
The shortlist
| Model | Tier | Weights | Context | Input / 1M | Output / 1M | Inputs |
|---|---|---|---|---|---|---|
| GPT-6 AstraOpenAI | Frontier | Closed | 1.05M | $10 | $50 | Text, Images |
| GPT-6.1 SolOpenAI | Advanced | Closed | 1.05M | $2 | $10 | Text, Images |
| GPT-6 LunaOpenAI | Efficient | Closed | 1.05M | $0.10 | $0.50 | Text, Images |
| Gemini 3.1 ProGoogle | Frontier | Closed | 1M | $2 | $12 | Text, Images, Video, Audio, PDF |
| DeepSeek-V4-ProDeepSeek | Frontier | Open weights | 1M | $1.32 | $3.96 | Text |
| Kimi K3Moonshot AI | Frontier | Open weights | 1M | $3 | $15 | Text, Images, Video |
| GLM-5.3Z.ai (Zhipu AI) | Frontier | Open weights | 1M | $1.40 | $4.40 | Text |
| Gemini 3.8 FlashGoogle | Advanced | Closed | 1M | $0.75 | $3.75 | Text, Images, Video, Audio, PDF |
| Gemini 3.6 FlashGoogle | Advanced | Closed | 1M | $0.75 | $3.75 | Text, Images, Video, Audio, PDF |
| Muse Spark 1.3Meta | Advanced | Closed | 1M | $1.25 | $4.25 | Text, Images, Video, PDF |
| DeepSeek-V4.1-FlashDeepSeek | Advanced | Open weights | 1M | $0.30 | $1.20 | Text, Images |
| MiniMax-M3MiniMax | Advanced | Open weights | 1M | $0.30 | $1.20 | Text, Images, Video |
| Gemini 3.5 Flash-LiteGoogle | Efficient | Closed | 1M | $0.30 | $2.50 | Text, Images, Video, Audio, PDF |
| Gemini 3.1 Flash-LiteGoogle | Efficient | Closed | 1M | $0.25 | $1.50 | Text, Images, Video, Audio, PDF |
| GLM-5.3-FlashZ.ai (Zhipu AI) | Efficient | Open weights | 1M | $0.15 | $0.50 | Text, Images, Video |
| Claude Fable 5.1Anthropic | Frontier | Closed | 1M | $10 | $50 | Text, Images, PDF |
| Claude Opus 5.5Anthropic | Frontier | Closed | 1M | $4 | $20 | Text, Images, PDF |
| Mistral Large 4Mistral AI | Frontier | Closed | 1M | $1.36 | $4.18 | Text, Images |
| Qwen3.8-MaxAlibaba (Qwen) | Frontier | Closed | 1M | $2 | $6 | Text, Images |
| Claude Sonnet 5.5Anthropic | Advanced | Closed | 1M | $2 | $10 | Text, Images, PDF |
| NVIDIA Nemotron 3 UltraNVIDIA | Advanced | Open weights | 1M | $0.50 | $2.20 | Text |
| Claude Haiku 5.5Anthropic | Efficient | Closed | 1M | $0.10 | $0.50 | Text, Images, PDF |
| Grok 4.3xAI | Efficient | Closed | 1M | $1.25 | $2.50 | Text, Images |
| Amazon Nova 2 LiteAmazon (AWS) | Efficient | Closed | 1M | — | — | Text, Images, Video, PDF |
| NVIDIA Nemotron 3.5 Lightning 30B-A3BNVIDIA | Compact | Open weights | 1M | $0.05 | $0.20 | Text |
| Grok 4.7xAI | Frontier | Closed | 500K | $2 | $6 | Text, Images |
| Qwen3.8-2.4T-A95BAlibaba (Qwen) | Frontier | Open weights | 256K | $2 | $6 | Text |
| Mistral Large 3Mistral AI | Advanced | Open weights | 256K | $0.50 | $1.50 | Text, Images |
| Mistral Medium 3.5Mistral AI | Advanced | Open weights | 256K | $1.50 | $7.50 | Text, Images |
| Mistral Small 4Mistral AI | Efficient | Open weights | 256K | $0.15 | $0.60 | Text, Images |
Ordered by context window. Tiers are our editorial grouping of provider positioning and public benchmark standings, not our own test results. How we rank.
How to choose
- Check the maximum output separately: a huge input window can still come with a modest output limit.
- Look for long-context surcharges in the price notes.
- Use the context window visualizer to see what fits.
Frequently asked questions
Which LLM has the largest context window?
The model at the top of this table has the largest advertised window among the current models we track. Very large windows exist mainly on a few models; most current frontier models offer around 1 million tokens.
Does a bigger context window make a model smarter?
No. It only lets the model see more text at once. Quality on long inputs depends on how the model was trained and evaluated for long context, so test with your own documents.
Other shortlists
Need hands-on help building it?
RAG, agents, fine-tuning or a production chatbot: describe the project and we'll suggest an approach and, if you want, partners who build it.
- Free and vendor-neutral: no model provider pays for placement.
- Your details go only to partners you agree to.