Context window

The definition

The text a model can read within one session. It is transient and ends with the session; a bigger window fits more into one session but does not carry the session forward.

The same words the glossary hub and the public data mirror carry; this page is the stable address for the term.

Where it shows up on this site

• Why AI forgets · ReferenceContext windows, sessions and what resets
• Open models · DirectoryQwen, DeepSeek, Mistral, Gemma, gpt-oss, Kimi, GLM, Meta's Muse Spark Open and more, read from their makers' own pages and graded on the same seven questions: the license, whether the weights download, whether the context window is stated, what it takes to run locally, whether the maker publishes a hosted price, whether your prompts train the maker's models, and whether you can fine-tune it and keep the result.
• Muse Spark Open license and weights · MetaMeta's open-weight model from Meta Superintelligence Labs, released 15 October 2026: 816B total and 42B active parameters with a 1M-token context window. It is Meta's newer open release than the Llama 4 models of April 2025. 3.5/7 on the open models tests.
• Kimi K3 license and weights · Moonshot AIMoonshot AI's open-weight native multimodal agentic model: 2.8T total and 104B active parameters with a 1-million-token context window, served on the Kimi Open Platform as kimi-k3. 4.5/7 on the open models tests.
• Gemma 4 license and weights · Google DeepMindGoogle DeepMind's family of open models in five sizes (E2B, E4B, 12B, 26B A4B and 31B), with a 256K context window on the medium models and Apache 2.0 licensing; the two largest also run on the Gemini API. 6/7 on the open models tests.
• RULER · MatureLong-context benchmark. A synthetic, configurable suite measuring the real usable context size of long-context models, not just their advertised context window.