Almanac
benchmark

PinchBench

benchmarkactiveprovisionalpinchbench-2e9533f8·2 events·first seen 14d ago

Aliases: PinchBench

Co-occurring entities

More like this (12)

Recent events (2)

7The Batch·14d ago·source ↗

Nvidia releases Nemotron 3 Super 120B-A12B open-weights model with hybrid Mamba-2/MoE architecture

Nvidia released Nemotron 3 Super 120B-A12B, an open-weights LLM with a hybrid Mamba-2/transformer/MoE architecture that activates only 12B parameters per token and supports up to 1 million token context. The model claims the fastest inference speed in its size class at 442 tokens/second and leads open-weights models on PinchBench agentic task evaluation, outperforming larger models including Kimi K2.5 (1T parameters). Nvidia is releasing weights, training data, and recipes under a permissive commercial license, and plans a $26B five-year investment in open-weights models — framed partly as a strategic response to Chinese labs building capable open-weights models on non-Nvidia hardware.

6arXiv · cs.AI·32h ago·source ↗

TokenPilot: Dual-granularity context management cuts LLM agent inference costs by up to 87%

TokenPilot is a cache-efficient context management framework for LLM agents that addresses the trade-off between token sparsity and prompt cache continuity. It combines Ingestion-Aware Compaction (global prefix stabilization) with Lifecycle-Aware Eviction (local segment offloading) to reduce inference costs by 56–87% across benchmarks while maintaining competitive task performance. The system is evaluated on PinchBench and Claw-Eval and has been integrated into the open-source LightMem2 library.