jegreene/Swift-1.5-Qwen3.8-27B-heretic-GSQ-RCO-GGUF
image-text-to-text model by jegreene
From the publisher
Model card & documentation
Source previewRead the publisher’s intended use, setup instructions, evaluations and limitations. The original model card is the source of truth.
This is a partially decensored build of UkisAI's Swift 1.5 Qwen3.8-27B, made with Heretic. It is packaged as a single-file llama.cpp GGUF and quantized the same way as UkisAI's own GSQ-RCO -mtp releases, using the same per-tensor allocations and the same imatrix. Three sizes are provided: IQ3S, IQ3XXS and IQ2S. It includes the MTP head, for speculative decoding, and a vision projector (mmproj) built from Swift's own vision tower. If you like this model, please donate (BTC): All three GGUFs are quantized from the same abliterated weights and share the one mmproj file. IQ3S is the recommended file. The two smaller ones trade quality for size: MTP acceptance figures. The IQ3XXS and IQ2S files have been checked tensor by tensor against UkisAI's allocations, but they have not been benchmarked or speed-tested. their own (non-abliterated) files at the same tiers: IQ3S 0.051, IQ3XXS 0.098, IQ2S 0.135. The abliteration's own 0.082 comes…
Read the full model card ↗ · Preview checked 2026-10-31T07:05:28.212Z
Inside the original model card — Document outline
- Swift 1.5 Qwen3.8-27B — Heretic (abliterated) · GSQ-RCO GGUF (IQ3S, IQ3XXS, IQ2S, all with MTP)
- Files
- How uncensored is it? Read this first
- Benchmarks
- Performance notes
- Usage (llama.cpp)
- Chat template notes
- How it was made
- Intended use and responsibility
- License
- Credits
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Links from the model card
References supplied by the publisher, not independently verified endorsements. Check the destination before downloading files or entering credentials.
Documentation belongs to its respective authors. Reported project/model license: other. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
Model overview
jegreene/Swift-1.5-Qwen3.8-27B-heretic-GSQ-RCO-GGUF is a image-text-to-text model repository published by jegreene on Hugging Face. The source reports the gguf library.
This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.
Model facts
- Task
- image-text-to-text
- Library
- gguf
- Recent downloads (30 days)
- 732,223
- Cumulative likes
- 27
- Hugging Face trending score
- 7
- Reported safetensors parameters
- Not reported
- Architecture
- Not reported
- License
- other
- Access
- Not gated by Hugging Face
- Created
- 2026-09-29T13:29:15.000Z
- Last modified
- 2026-10-27T22:16:19.000Z
Compatibility and lineage
Language coverage was not reported.
Reported base models: ukisai/Swift-1.5-Qwen3.8-27b
Parameter count is not a RAM/VRAM requirement. Check precision, quantization, context length and runtime compatibility in the model card; no hardware or API-cost claim is inferred here.
Use and evaluate
Check license terms and access requirements first. Review the model card’s documented loading instructions, evaluations and safety limitations. Benchmark scores are not imported by this directory, and popularity is not an accuracy ranking.
No model weights are downloaded or executed by AltAPIs. Never enable remote model code without reviewing it.
Source and freshness
Source: Hugging Face Hub. Metadata observed 2026-11-01T06:51:30.840Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.