---
format: "aidr-story-markdown/v1"
id: "f0b4d040fb638191a1a7fe198e176dd5632495c0a7511c9a75dce42cceabb277"
canonical_url: "https://aidr.today/f0b4d040?lang=en"
title: "Qwen3.8-27B Model Runs on 8GB RAM via 1-Bit Quants Retaining 77% Accuracy"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-08-19T18:53:11.000Z"
category: "Models"
topics: ["qwen","alibaba","open-source","inference","coding"]
source_urls: ["https://huggingnews.com/ai/update-qwen38-27b-model-runs-on-8gb-ram-via-1-bit-quants-retaining-77per-3c841aaa","https://x.com/danielhanchen/status/2090104316619268381","https://x.com/danielhanchen/status/2090119165055324518","https://x.com/tinkerapi/status/2090129004091306401","https://x.com/Alibaba_Qwen/status/2089922175507865795","https://x.com/Alibaba_Qwen/status/2089919106522976337"]
summary: "The AI optimization firm Unsloth released new GGUF files and quantized weights that enable Alibaba's Qwen3.8-27B model to operate on consumer hardware with minimal memory. The 1-bit quantized versions can run on systems with 8GB of RAM while retaining approximately 77% of the accuracy of the BF16 version. Additionally, the Unsloth Dynamic v3 GGUFs provide a 10% increase in top-1% accuracy, validated by a new metric called Divergence-300 which uses 300 unseen examples from DeepSWE and Terminal Bench to measure greedy accuracy across 32 tokens. Tinker launched the Qwen3.8-27B model on its platform, providing access to native multimodal capabilities for images and video. The model is designed for professional work, research, and long-horizon agentic tasks, and features flexible thinking control alongside improved coding performance."
---

# Qwen3\.8\-27B Model Runs on 8GB RAM via 1\-Bit Quants Retaining 77% Accuracy

> [Open the canonical story](<https://aidr.today/f0b4d040?lang=en>)

**Published:** 2026-08-19T18:53:11.000Z
**Category:** Models
**Topics:** qwen, alibaba, open\-source, inference, coding

## Summary

The AI optimization firm Unsloth released new GGUF files and quantized weights that enable Alibaba's Qwen3\.8\-27B model to operate on consumer hardware with minimal memory\. The 1\-bit quantized versions can run on systems with 8GB of RAM while retaining approximately 77% of the accuracy of the BF16 version\. Additionally, the Unsloth Dynamic v3 GGUFs provide a 10% increase in top\-1% accuracy, validated by a new metric called Divergence\-300 which uses 300 unseen examples from DeepSWE and Terminal Bench to measure greedy accuracy across 32 tokens\. Tinker launched the Qwen3\.8\-27B model on its platform, providing access to native multimodal capabilities for images and video\. The model is designed for professional work, research, and long\-horizon agentic tasks, and features flexible thinking control alongside improved coding performance\.

## Sources

- [Story source](<https://huggingnews.com/ai/update-qwen38-27b-model-runs-on-8gb-ram-via-1-bit-quants-retaining-77per-3c841aaa>)
- [Story source](<https://x.com/danielhanchen/status/2090104316619268381>)
- [Supporting source](<https://x.com/danielhanchen/status/2090119165055324518>)
- [Supporting source](<https://x.com/tinkerapi/status/2090129004091306401>)
- [Story source](<https://x.com/Alibaba_Qwen/status/2089922175507865795>)
- [Supporting source](<https://x.com/Alibaba_Qwen/status/2089919106522976337>)