Benchmark: DFlash1 vs DFlash2 Speed on Qwen Models
AlpinDale · x · 2026-08-20
A developer benchmarked the performance of DFlash1 and DFlash2 speculation decoding techniques on Qwen models:
- Speed: DFlash1 appears faster for Role Play (RP) tasks compared to DFlash2.
- Constraints: DFlash2 is currently limited to Qwen3.8 and caps speculative tokens at a maximum of 7, whereas DFlash1 supports higher limits (up to 15).
- Setup: Comparison ran on Qwen3.6-27B-FP8 (with DFlash1) vs Qwen3.8-27B-FP8 (with DFlash2).
More from Infra
- Large data centers employ only ~50-150 permanent on-site staff, Microsoft says — brezina · 2026-10-16
- Prediction: Opus 5.5-level models on laptops by 2027, backed by Qwen 27B data — cephaloform · 2026-10-16
- Oracle down 12% in a week, nearly 50% off June high as OpenAI exposure spooks market — alvelda · 2026-10-16
- US data center moratoriums top 400 — an AI bull explains why they're actually good for AI — kevinsxu · 2026-10-16
- Microsoft Research unveils Magnetar, the fastest zkVM at 10-39x speedups and post-quantum security — jedisct1 · 2026-10-16
- Bonsai 2 2B hits 120 tok/s on iPhone 18 Pro, outpaces Qwen 3.5 2B at half the memory — cephaloform · 2026-10-16