Escha Labs Inc. released Escha-W2, a 2-bit quantized build of Qwen3.8-27B that compresses the 27-billion-parameter model to 10.15GB, enabling 128k context on a single 24GB GPU while matching FP8 quali…
Ornith AI released Ornith-1.0-397B, a 397-billion-parameter mixture-of-experts model with 262K context, on 2026-06-25, now superseded by Ornith-1.5-397B. The model self-reports coding scores of 77.5 o…
Ornith AI released Ornith-1.0-35B, a 35B-parameter mixture-of-experts model with 3B active parameters per token, on June 25, 2026, under an MIT license on HuggingFace, featuring a 262K context window.…
A new arXiv paper (2608.18795v1) quantifies why majority voting among LLM samples can backfire on hard questions, introducing a pluralistic agreement index Gamma decomposed into a mechanical preferenc…
A new arXiv paper (2607.17531v1) introduces a fixed-pool diagnostic framework for test-time collaboration in LLMs, decomposing net gains into recoverable mass, verification-signal coverage, conditiona…
A new method called HALO (Hybrid Adaptive Latent Reasoning) improves frozen pretrained language models with adaptive extra computation, achieving the best overall average on MMLU-Pro and GPQA-Diamond …
A new study testing 67 frontier language models from 21 providers found that combining multiple models rarely outperforms the single best model, with gains capped by a 'co-failure ceiling' where all m…
Researchers propose SuperThoughts, a method that compresses pairs of consecutive Chain-of-Thought tokens into single latent representations and decodes two tokens per step, doubling inference throughp…
Researchers at Duke University introduced Reward-Variance Policy Optimization (RVPO), a risk-sensitive alignment method that penalizes inter-reward variance to prevent language models from neglecting …