For privacy reasons a browser tells us at most "≥ 8 GB RAM, 8 cores" - same reading whether you have 8 GB or 128 GB. It has no idea how much RAM is free right now, which apps are open, or whether you have a GPU.
It reads your exact RAM, GPU model and VRAM, free memory right now, and picks the sharpest quant that still fits. Every model page lights up precisely for your rig.
The app is a full local runtime - no API keys, no subscription, everything runs on your machine. Click any model on this site and start a conversation in seconds.
Get the free app →This is a rough estimate. Install the free app - we'll show exact numbers.
Reading real hardware from your app right now. Numbers below are exact.
Below is the per-quantization compatibility for this model.
curl -H "Authorization: Bearer $ABL_KEY" \
"https://abliteration.org/api/v1/models/zetasepic%2FQwen2.5-32B-Instruct-abliterated-v2-GGUF" No specific extraction method could be identified for this model. The producer either did not document it or used a proprietary pipeline.
What is a refusal direction? →| Benchmark | Score | Source |
|---|---|---|
| Entertainment | 1.9 | UGI |
| Hazardous | 3.5 | UGI |
| Natural Intelligence | 21.93 | UGI |
| Political lean | -12.1% | UGI |
| Sensitive-Info | 27.63 | UGI |
| SocPol | 3.2 | UGI |
| UGI | 44.25 | UGI |
| Willingness (10) | 7.8 | UGI |
| W10-Adherence | 6.5 | UGI |
| W10-Direct | 9 | UGI |
| Writing | 32.89 | UGI |
| Benchmark | Score | Source |
|---|---|---|
| BBH average | 0.6076015047347256 | OpenLLM-v2 |
| IFEval instruct | 0.8633093525179856 | OpenLLM-v2 |
| IFEval-Prompt | 0.8059149722735675 | OpenLLM-v2 |
| MATH lvl 5 | 0 | OpenLLM-v2 |
| MMLU-Pro | 0.566655585106383 | OpenLLM-v2 |
This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.
The same weights this author released in different packaging. Pick the format that matches your runtime.
| Qwen2.5-32B-Instruct-abliterated-v2-Q8_0.gguf | 32.4 GB | 80790cc8 | download |
| Qwen2.5-32B-Instruct-abliterated-v2-Q6_K.gguf | 25.0 GB | 41f96e21 | download |
| Qwen2.5-32B-Instruct-abliterated-v2-Q4_K_M.gguf | 18.5 GB | 1984fb0c | download |
| Qwen2.5-32B-Instruct-abliterated-v2-IQ4_XS.gguf | 16.6 GB | 24f5d098 | download |
| Qwen2.5-32B-Instruct-abliterated-v2-Q3_K_M.gguf | 14.8 GB | e6c17be9 | download |
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen2.5-32B-Instruct/blob/main/LICENSE
language:
Abliterated version of Qwen/Qwen2.5-32B-Instruct, utilizing code from refusal_direction.
For more information about the Abliterated technique, refer to this article and check out @FailSpy.
The author's README evolved over time. Click a version to see its content at that point.
Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.