Ingot quantizes Qwen 27B to FP8 in one week, beating Alibaba's own quantization
const_reborn · x · 2026-10-22
Ingot-I quantized Qwen 27B into FP8 in one week, outperforming Alibaba's official quantization: 56.2%/60.7% on Terminal-Bench 2.1 vs 52.8% for official FP8, and 80.8% on GPQA Diamond. Next targets are 2-bit quantization and larger GLM models.
More from Models
- Daniel Rock: New frontier models react far more aggressively to context instructions — danielrock · 2026-10-22
- Researchers credit OpenAI's Daybreak program with GPT-5.6 access for Chrome, GitLab exploits — Eric_Wallace_ · 2026-10-22
- Zeroing 7 neurons cuts Qwen3-8B sycophancy from 76% to 0% in Opus-assisted experiment — Sauers_ · 2026-10-22
- Zeroing 6 neurons cuts Qwen3-8B sycophancy from 44% to 0%, interpretability charts show — Sauers_ · 2026-10-22
- User confused as Claude Code fast mode burns API credits instead of 20x subscription — RileyRalmuto · 2026-10-22
- Flint MoE model lands in llama.cpp: 39.8B total params, only ~2.5B active per token — jacek2023 · 2026-10-22