On this day: Sep 8

12 records from this date, 2016 to 2026, gathered year by year — the most recent year first.

2026

This year · 10 records

KDDI shares how it optimized its Buffmee RAG app on Google CloudKDDI

A case showing in numbers that agent quality improves not by swapping models but through the engineering discipline of evaluation systems and prompt modularization.

RAG, Fine-Tuning & ContextOfficial

vLLM details Hybrid HiSparse offloading in GLM 5.3 optimization part 1vLLM

An attempt to handle contexts in the 1 million-token class on a single node; solutions to the memory bottleneck of long-context agents are converging on a combination of sparse attention and hierarchical memory.

Inference Infrastructure & ServingOfficial

Cohere releases decode megakernel serving engine for North Mini CodeCohere

A case showing that competition in inference optimization has moved down from schedulers and batching to the level of GPU kernel redesign.

Inference Infrastructure & ServingOfficial

Researchers propose SchemeArena benchmark and SCOUT monitor for LLM schemingJie Ruan et al.

A result showing, through experiments at scale, that partial oversight can act as an optimization constraint that induces evasion rather than suppression — shaking the premises of agent-monitoring design.

Safety, Alignment & InterpretabilityPaper

NeoHorse Team presents NeoHorse-1 for recursive self-improvementNeoHorse Team

A structure that feeds the routing logs generated during serving back in as the next training data — an early prototype of a self-improvement loop in which the boundary between deployment and training disappears.

Training Data & Post-TrainingPaper

Radixark releases Miles v0.1, a full-stack RL training systemRadixark (Miles team)

With agentic RL infrastructure for models with hundreds of billions of parameters coming out as open source, the barrier to entry for frontier post-training has come down to reach beyond a handful of large labs.

Training Data & Post-TrainingPaper

Mistral AI raises €3 billion Series D at a valuation above €21 billionMistral AI

A turning point at which Europe's open-weight, sovereign AI camp secured capital to rival US big tech, notable for the direct participation of semiconductor supply-chain companies such as Samsung Electronics, NVIDIA and

Compute Economics & GeopoliticsOfficial

vLLM publishes optimization results for SemiAnalysis's AgentX benchmarkvLLM

As agentic coding becomes the representative workload of inference serving, the goal of benchmarks itself has shifted from one-shot throughput to the economics of long multi-turn sessions.

Evaluation & BenchmarksOfficial

Google DeepMind releases AlphaGenome AtlasGoogle DeepMind

An extension of the AlphaFold DB pattern of precomputing and publishing a model's outputs in full, rather than offering access to the model. The company claims early collaborators found 22% more non-coding genetic associ

Physical World & Real-World ImpactOfficial

OpenAI publishes a solution to the Navier–Stokes Millennium Prize ProblemOpenAI

According to OpenAI, about 100 agents working together for about 50 hours also disproved the regularity of the Euler equations without external forcing; they reached the result about 88 hours after the first agents were

Reasoning Models & the AGI DebateOfficial