Tag
A user shares benchmarks showing that raising the --adapt-every parameter from 1 to 6 in Strata boosts token generation speed by 9-55% across multiple Qwen 3.8 Flash Next prompts, with the largest gains on long-context code descriptions.
Inco AI announces DFlash 2, a next-generation decoding technique that achieves up to 4.6× speed increase for Qwen3.8-27B on M5 Max MacBook Pro, improving efficiency without altering output.
The lab behind LLaDA2.2 released a diffusion model benchmarked against its own autoregressive model, showing diffusion lags on general knowledge and coding but wins on speed and agent tasks, providing a clean tradeoff data point.