Unverified50% confidenceBenchmarkExact time
Under a representative configuration with block size B=8, BD achieves an evaluation speedup of about 1.43x.
1
Sources
50%
Confidence
Long-term
Relevance
10/22/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedControlled measurements show that branch pruning brings a 1.426x speedup in step time.67% similarUnverifiedIn the B300 EP8 decoding scenario, Chord achieves a 2.15× speedup compared to the untuned default implementation.66% similarUnverifiedIn the LongLive benchmark test, Speculative Forcing matches the quality of a four-step generator, requires on average only about 1.7 denoising steps per token, runs more than 1.5 times faster, and reaches generation frame rates of 41.7 and 37.8 fps.62% similarUnverifiedMixtral-8x7B achieves performance close to that of a 70B dense model with a total of 47B parameters, while offering faster inference speed62% similarUnverifiedParameter count does not reliably predict indexing speed: EmbeddingGemma-300M indexes more slowly than the larger BGE-M3 (315.0 vs 433.0 chunks per second).62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1084307API
curl https://kongchang.com/api/v1/knowledge/claims/1084307MCP
get_claim(id=1084307)