Open-source Hemmingway-1: a 27B creative-writing model that beat GPT-5.5 in blind tests
Lukinator6446 · reddit · 2026-09-21
Altworld open-sourced Hemmingway-1, a 27B model (based on Qwen3.8) trained exclusively for creative writing, motivated by ChatGPT stories sounding like 'a polite intern wrote them.' It scores 1330 on EQ-Bench 4 and ranked first in the team's blind writing tests, ahead of GPT-5.5 and other frontier models. It runs on a single 24GB GPU, is free to try on their site, and is explicitly not for math or code. Weights are on Hugging Face.
Related event: Two-person lab Altworld open-sources writing model Hemmingway-1(7 posts)→
More from Models
- Why Is Inkling 2 Small Trained on So Few Tokens Yet Still Competitive? — georgejrjrjr · 2026-11-12
- OpenAI's five biggest failures laid out: pretraining decline, alignment fears, enterprise lost to Claude — teortaxesTex · 2026-11-12
- Fable 5.2 Hits 41.3% on Remote Labor Index, Up From 2.5% a Year Ago — alliekmiller · 2026-11-12
- Beaker automates MIT's weight-space programming: one-neuron edit lifts MATH-500 from 7% to 37% for under $4 — VictorSanh · 2026-11-12
- Researcher says Anthropic dodged questions on Mythos false positive rate; 37,966 findings, 2 exploited — OwariDa · 2026-11-12
- Grok spotted building meeting "Listen to Conversation" feature with live transcript — nima_owji · 2026-11-12