Havuz

Can AI agents rebuild websites through browser exploration?

SWE-sweep, agent'ların büyük kod tabanlarındaki konumu ve türü önceden belirtilmeyen birden fazla bug'ı bulup düzeltebilmesini ölçen bir benchmark'tır.

Neden yükseliyor · 1 günlük, 33 yıldızlı yeni bir Facebook Research benchmark'ı olması ve LLM agent değerlendirmesine yönelik güncel bir soruyu ele alması dikkat çekebilir. Ancak yükselişin kesin nedeni bilinmiyor; son 28 günlük commit ve farklı yazar verisi ölçülmemiş.

[Public preview] Externally scored agentic ML research benchmark: 60 tasks, real competition ground truth. Open protocol, operated evaluatio

A decision library for running local LLMs on multi-GPU machines without NVLink: parallelism, communication, quantization, speculative decodi

HEWN 2.0 2026: AI Output Router for Precision Summaries & Polished Code

Compute-for-memory research lab for running larger LLMs on memory-constrained consumer GPUs. Official website exvram.pro

Real-world percentile benchmark for LLM inference servers: 1% A/B resolution, offline HTML reports. I am now on X also: https://x.com/bkuype

Low-memory C++20 LLM inference runtime for quantized GGUF models on CPU — mmap-backed weights, paged KV cache, Q4/Q8 kernels, and reproducib

Semantic cache strategy bench: replay recorded request traces through candidate caching policies and measure which one actually pays off.

Iterative agent harness improvement: run a coding agent on a hard task, generate the reusable tooling it was missing, qualify it, and replay

Dual-engine (llama.cpp + vLLM) LLM benchmarking pipeline for GGUF & safetensors on NVIDIA GPUs — speed, quality, live dashboard, publishable

A lightweight, general-purpose harness for tool-using LLM agents, fair benchmark evaluation, harness baselines, and personal assistant workf

A pure-Python structured memory benchmark for multi-agent LLM systems — context graph vs vector RAG vs raw history dump, five scenarios, 18

Dashboard for InferenceX™, Open Source Continuous Inference | InferenceX 仪表板

LLM benchmark results for THOR forensic finding triage quality

Auditable context capsules for LLM handoffs, coding agents, and OpenCode MCP workflows.

Open benchmark for expert collapse and routing efficiency in sparse MoE LLMs

Automated auditing pipeline for LLM and agent benchmarks — surfaces task ambiguity, environment conflicts, and evaluation bugs.