Unverified50% confidenceFactExact time
The VRAM occupied by the context window is calculated separately from the model weights, and KV cache VRAM usage grows linearly with context length.
1
Sources
50%
Confidence
Long-term
Relevance
8/14/2026
First Seen
Sources
看懂GGUF量化：Q4_K_M与Q4_K_S的真正区别
Related Claims
VerifiedThe autoregressive generation process of reasoning models is strictly sequential; KV Cache VRAM usage grows linearly with the length of the thinking token sequence, making memory bandwidth the performance bottleneck.74% similarUnverifiedThe available memory and VRAM of a local model directly determine the obtainable context window length, because inference requires maintaining a KV Cache in VRAM.73% similarUnverifiedAutoregressive models use KV Cache to cache the Key and Value matrices of historical tokens, with memory usage growing linearly with sequence length.72% similarUnverified条件化历史帧方法随上下文增长计算量呈二次方增长，而状态空间模型方法以线性复杂度维护隐式场景状态72% similarUnverifiedThe memory of LLMs depends on the context window, which is the maximum number of tokens the model can process in a single inference.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/745193API
curl https://kongchang.com/api/v1/knowledge/claims/745193MCP
get_claim(id=745193)