Unverified50% confidenceFactExact time
Baidu's UnlimitedOcr is a 3B-parameter OCR model with a context length of 32,768 tokens, built on DeepSeek OCR, replacing all decoder attention layers with Reference Sliding Window Attention (R-SWA), keeping the KV cache constant, and can transcribe dozens of pages in a single forward pass.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/15/2026
First Seen
Valid until: 1/13/2027
Sources
Related Entities
Related Claims
UnverifiedReference Sliding Window Attention (R-SWA) introduces a fixed reference position on top of the sliding window, keeping the KV cache at a constant size during decoding while not sacrificing cross-page context consistency.68% similarUnverifiedO_DIRECT is a Linux file open flag that bypasses the page cache and transfers data directly from disk to user-space buffers, letting a program fully control memory usage and avoid double-copy overhead.64% similarVerifiedPrompt caching saves pre-computed KV matrices on the server side, reducing the processing of repeated prefixes from O(n²) to O(1) cache reads, enabling up to 90% compression in latency and cost.64% similarUnverifiedIn large model API call scenarios, cache hit rate refers to the reuse ratio of the KV Cache (key-value cache), where the model can directly read the attention key-value pairs of already-processed tokens from the cache.63% similarUnverifiedKV Cache memory usage is proportional to the number of layers × number of attention heads × context length × precision; GQA reduces cache size by having multiple query heads share the same set of Key-Value heads, and this design has been adopted by the Llama 2/3 and Qwen series.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1039066API
curl https://kongchang.com/api/v1/knowledge/claims/1039066MCP
get_claim(id=1039066)