HF model

dahara1 / shisa-v2.1-qwen3-8b-UD-japanese-imatrix

A GGUF‑quantized shisa‑v2.1‑qwen3‑8b model built with Unsloth Dynamic 2.0, community‑patched Qwen3 settings to reduce malfunctions, a larger imatrix for stronger Japanese performance, and a 40K maximum context length.

Description in other languages

Statistics

Statistics as of Nov 16, 2026

Hugging Face downloads (last 30 days)
2,000
Hugging Face likes
5
Listed
Dec 23, 2025

Downloads (30-day window) over time

02.5k5k7.5k10kNovSepJulMayMarJan 20262,000
Show as a table
MonthDL (30d)
Dec 20258,000
Jan 20261,000
Feb 20262,000
Mar 2026—
Apr 20262,000
May 2026—
Jun 2026—
Jul 20262,000
Aug 2026—
Sep 20263,000
Oct 20261,000
Nov 20262,000

Downloads over the 30 days before the last snapshot of each month.

Popular in the same category