Nous Research · Evaluation & Benchmarks
Atropos
Async-first LLM RL environments framework for collecting and evaluating model trajectories across diverse environments (GSM8K, MMLU, multi-turn games, code execution, multimodal). Demonstrates 4.6x improvement on parallel function calling and integrates with Axolotl, Tinker, vLLM, and SGLang.
Open the source ↗Similar Developer Tools releases
- AWS Emulator
AWS · Agent-built AWS emulator · Week 47, 2026
- UAEF (Universal Agent Evaluation Framework)
AWS Machine Learning · Agent Eval Framework · Week 47, 2026
- Token Pooling-Aware Training for Multi-Vector Embedding Models
Hugging Face (Tom Aarsen) · Pooling-aware embeddings · Week 47, 2026
- Gemini Decisions API (Gemini Decision 1)
Google DeepMind · Multimodal Decision API · Week 47, 2026
- WorldFoundry
OpenEnvision (Juanxi Tian, Jiadong Hou, Baiang Li, Xin Lin, Zhengbo Jiao, Weijia Zhang, Yeying Jin, Truong Nguyen) · World model infrastructure · Week 47, 2026
- verifiers
Prime Intellect · LLM RL environments · Week 45, 2026
More from Nous Research
- Arete-27B
Nous Research · Propaganda-reduced fine-tune · Week 46, 2026
- Hermes Agent
Nous Research · Agent Framework · Week 18, 2026
- Atropos
Nous Research · Agent Framework · Week 11, 2026
- NousCoder-14B
Nous Research · Code Generation · Week 2, 2026
Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.