Part 2 · The Harness Around the Model

Context Window: AI's Working Memory

Window composition visualization; drag to simulate overflow; compare capacities of mainstream models

THE QUESTION THIS PAGE ANSWERS

ANSWER FIRST

What is the key idea behind “Context Window: AI's Working Memory”?

Window composition visualization; drag to simulate overflow; compare capacities of mainstream models

DECISION RULE

Make the claim earn its place. Use this page as a decision aid, not a definition to memorize. Connect the idea to one real task, one observable result, and one failure that would change your mind.

TRY NEXT

Write one question you could answer with evidence after trying this idea.

WATCH FOR

A conclusion that sounds complete but leaves the key assumption untested.

What is the context window?

The context window is everything the model can see in a single call. Anything beyond the window is completely gone — not blurry, not vague, simply invisible.

Mainstream model context windows compared (2025)
256K
Kimi K2.5
~200K Chinese characters
The entire Romance of the Three Kingdoms
256K
Qwen3.6-Plus
Native 256K
Expandable to 1M
400K
GPT-5
~320K Chinese characters
Available for Pro users
1M
Claude Sonnet 4.6
~800K Chinese characters
Dream of the Red Chamber ×5
1M
Gemini 2.5 Pro
~800K Chinese characters
2M coming soon
💰 Larger window = higher cost:
LLMs are billed by input + output Token count. Uploading a 100-page PDF (≈50K Tokens) adds roughly $0.07 per conversation (GPT-5 reference price).
After multiple turns, conversation history also accumulates and fills the window. Context is a finite, priced resource.
Interactive demo · Drag the slider to experience overflow

Context breakdown for one request (128K window)

0K / 128K
System PromptChat HistoryUploaded DocCurrent QueryReserved for Reply
System Prompt (10K)
Chat History (20K)
Uploaded Doc (variable)
Current Query (5K)
Reserved Reply (10K)
Window is sufficient — 83K Tokens remaining
Simulated upload size (drag to experience overflow) 0K Tokens
📏 The window is a finite resource. Using it costs money; exceeding it causes truncation. See the next page for three strategies when the window overflows.

How “What is the context window” changes an answer

“The context window is everything the model can see in a single call.” shows that a model does not process the “word count” we see. It processes Token pieces. Tokenization affects input length, how much context fits, and how much computation a request consumes.

Length, information, and context are different

As “The context window is everything the model can see in a single call.” grows, separate three questions: how many Tokens the text becomes, which pieces can change the current decision, and whether older material has fallen outside the context window. Removing repetition is often more useful than simply making the window larger.

Keep what can change the decision

Use “The context window is everything the model can see in a single call.” as an A/B test: keep the same question while removing repeated background, compressing format, and trimming irrelevant history. Compare answer quality, latency, and Token count.

From “What is the context window” to “Mainstream model context windows compared (2025)”

“What is the context window” grounds the problem in “The context window is everything the model can see in a single call. Anything beyond the window is completely gone — not blurry, not vague, simply invisible”. “Mainstream model context windows compared (2025)” then moves it toward “256K Kimi K2.5 ~200K Chinese characters The entire Romance of the Three Kingdoms 256K Qwen3.6-Plus Native 256K Expandable to 1M 400K GPT-5 ~320K Chinese characters Available for Pro users 1M Claude Sonnet 4.6 ~…”. Together, they show that the lesson is not just a conclusion to remember, but a claim with conditions.

Carry the judgment into the next situation

For long text, keep what can change the conclusion before compressing format and history. A larger context is worth its cost only when the added information is useful.

  • “What is the context window”: The context window is everything the model can see in a single call. Anything beyond the window is completely gone — not blurry, not vague, simply invisible
  • “Mainstream model context windows compared (2025)”: 256K Kimi K2.5 ~200K Chinese characters The entire Romance of the Three Kingdoms 256K Qwen3.6-Plus Native 256K Expandable to 1M 400K GPT-5 ~320K Chinese characters Available for Pro users 1M Claude Sonnet 4.6 ~…
  • “The closing point”: The context window is everything the model can see in a single call. Anything beyond the window is completely gone — not blurry, not vague, simply invisible

The final “The closing point” brings the discussion to “The context window is everything the model can see in a single call. Anything beyond the window is completely gone — not blurry, not vague, simply invisible”. The useful thing to carry forward is knowing which judgments must be revisited when input, scale, or risk changes.

Mark as learned Your reading progress updates automatically
← PreviousNext →

Keep reading

The next useful article in the thread.

ARTICLE DISCUSSION

Leave one useful thought here.

Keep the idea that clicked, the question that stayed open, or a small note for the next learner.

Discussing Context Window: AI's Working Memory The Harness Around the Model
3discussionsArticle discussion · synced with the Circle
View in the learning circle
AM
Asha MorganContent editor
INSIGHTField note

I turned one judgment from this article into a small experiment I could run today. Knowing what to observe next is more useful than simply remembering the conclusion.

ARTICLE DISCUSSION7 helpful
LH
Lin HarperIndie developer
INSIGHTInsight

After reading this, I first looked for the conditions behind the idea instead of copying the method into a project. That order made the later trade-offs much clearer.

ARTICLE DISCUSSION5 helpful
KM
Kiki MooreProduct operations
QUESTIONQuestion

When this judgment reaches real work, which constraint should be added first? I am curious which step matters most between reading and the first practical attempt.

ARTICLE DISCUSSION4 helpful