Unverified50% confidenceFactExact time
CacheGraft is a retrieval-augmented generation method based on the Cache-to-Cache (C2C) fusion framework, which in an offline stage encodes each passage in the corpus into a fused KV cache representation between a large model and a small model, and directly injects it into the small model at query time without calling the large model.
1
Sources
50%
Confidence
Long-term
Relevance
11/6/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedBatching is a core means by which model serving improves throughput; continuous batching allows requests to join or leave the batch dynamically during generation, and paged KV cache management can reduce VRAM fragmentation.72% similarUnverifiedPrompt Cache本质上是大模型推理架构中KV Cache机制的服务端延伸,缓存Transformer中每个token生成的Key和Value矩阵72% similarVerifiedThe caching mechanism is technically commonly referred to as KV Cache or Prompt Caching69% similarUnverifiedContext Caching is a billing optimization mechanism provided by model vendors; when multiple requests share the same long prefix, cache breakpoints can be marked to reuse the already-computed KV Cache, billed at a lower unit price and reducing first-token latency.69% similarUnverifiedPrompt Caching is a mechanism provided by large-model service providers such as the Anthropic API: when the request prefix is exactly the same as that of a previous request, the server reuses the already-computed KV Cache, the cache-hit portion is billed at 0.1 times the input price, and the first cache write is billed at 1.25 times.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1141592API
curl https://kongchang.com/api/v1/knowledge/claims/1141592MCP
get_claim(id=1141592)