Unverified50% confidenceBenchmarkExact time
In evaluations across seven benchmarks, D3D achieves an average decoding speedup of 5.00x with Qwen3-8B as the target model.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/5/2026
First Seen
Valid until: 2/3/2027
Sources
Related Entities
Related Claims
UnverifiedUnder the same settings, DFlash achieves average decoding speedups of 4.03x and 3.67x with Qwen3-8B and Qwen3-32B as target models, respectively.79% similarUnverifiedOn the GSM8K benchmark, D3D reaches a speedup of 7.48x with Qwen3-32B as the target model.78% similarUnverifiedOn three benchmarks with Qwen3-8B as the target model, D3D still retains over 99% of the average acceptance length after pruning 5 of 36 layers.75% similarUnverified6.14x is the highest speedup measured on the specific model Qwen3-8B; the report does not provide complete curves across models or tasks, so it should be understood as a peak value under a specific configuration rather than a universal end-to-end throughput improvement.74% similarUnverifiedIn actual tests of 8-bit and 4-bit (OQ4E) quantization of the same Qwen3 27B model based on the MLX framework, each won 5 out of 10 accuracy tests, resulting in a tie.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1137902API
curl https://kongchang.com/api/v1/knowledge/claims/1137902MCP
get_claim(id=1137902)