AllFeed, trending, and board
FeedLatest news
TrendingOpen source and releases
BoardBlogs and showcases
TagsBrowse by topic

#speculative-decoding

The latest AI and developer news about #speculative-decoding, with the original source and a short summary.

AI Briefing

Privacy Policy Terms of Service Contact About Sources RSS Support us

Feed
Trending
Tags
Settings

Liquid AI Releases LFM2.5-VL-DSpark: An Experimental Draft Model for Faster Vision-Language Inference

·2026.09.24 23:00

Optimizing Gemma 4 Serving with vLLM on Amazon EKS [Part 2: Throughput and Latency SLOs on a Single GPU]

·2026.09.10 09:00

vLLM Publishes Performance Benchmarks for 5 Speculative Decoding Methods on AMD GPUs

·2026.09.07 18:00

3.85x Inference Speedup via Diffusion Draft Generation and AR Verification Based on GLM-OCR

·1·2026.09.03 18:00

The Efficient Frontier of LLM Inference (6-minute read)

·2·2026.09.02 09:00

vLLM v0.28.0 Released with Major Optimizations for Kimi-K3 and DeepSeek V4

·2026.08.31 09:00

LFM2.5 Inference Accelerated 3.2x with DSpark

·2026.08.21 01:00

DFlash 2: Maintaining Parallel Drafting

·2026.08.20 05:00

Arbitrage: Efficient Inference via Speculative Decoding Considering Model Superiority

·2026.08.07 09:00

DiffusionGemma Achieves 1,500 Tokens/s

·2026.08.06 15:00

FuriosaAI Presents Full-Stack Software Efficiency at ICML 2026

·2026.08.06 09:00

Fast Gemma's Proven Inference Optimization Recipe (7 min read)

·2026.08.04 09:00

Building Abundant Intelligence

·1·2026.08.01 00:00

llama.cpp Changes Default Loading of MTP Tensors

·2026.07.30 03:00

llama.cpp Adds Support for DSpark Speculative Decoding

·2026.07.28 20:00

How to Build an Ultra-Fast API for GLM-5.2

·2026.07.27 09:00

LG AI Research: LLM Efficiency Research Through the Lens of ACL 2024

·2026.07.16 09:00

llama.cpp adds support for Tencent's Hy3 model

·2026.07.14 11:00

Show HN: Reame – a CPU inference server that gets faster the more it runs

·2026.07.12 01:00

Hardware Aware Dynamic Speculative Decoding

·2026.07.11 01:00

BlockPilot: Adaptive Block Size Prediction to Speed Up Inference

·2026.07.03 23:00

DeepSeek Open-Sources New Framework 'DSpark' That Boosts LLM Inference Speed by Up to 85%

·2026.06.30 09:00

DeepSeek releases DeepSpec for speculative decoding

·2026.06.28 23:00

DSpark: Speculative Decoding Framework Combining Semi-Autoregressive Generation and Confidence Scheduling Unveiled

·2026.06.28 09:00

1
2
3
Next

1
2
3
Next

#speculative-decoding | AI Briefing