greglechin/Swift-1.5-Qwen3.8-27B-Uncensored-GPTQ-Int4-sym-G128-MTP-BF16
image-text-to-text model by greglechin
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.
GPTQ-Int4 quantization of ajgazin/Swift-1.5-Qwen3.8-27B-Uncensored-MTP, built for Intel Arc / vLLM XPU with the MTP head preserved in BF16 so native speculative decoding still works. The recipe is a reproduction of kernelogic/Qwen3.8-27B-Uncensored-GPTQ-Int4-sym-G128-MTP-BF16, applied to the Swift lineage instead of the JonathanColetti one. Same quantizer version, same config, same MTP handling. Quantizing this model naively breaks speculative decoding. The 15 mtp. tensors are the draft head; if they get quantized along with everything else, draft acceptance collapses and you lose roughly half your decode speed. The fix is one line of quantize config: The result has 400 quantized weight tensors (I32) + 15 preserved BF16 MTP tensors. There is a second, less obvious requirement. transformers' Qwen35 declares keystoignoreonloadunexpected = [r"^mtp."], so the draft tensors are dropped at load and never become modules — the dynamic exclusion on its own emits a checkpoint with no draft head at all, silently. What writes them…
Read the full model card ↗ · Preview checked 2026-09-25T07:36:41.341Z
Inside the original model card — Document outline
- Swift-1.5-Qwen3.8-27B-Uncensored-GPTQ-Int4-sym-G128-MTP-BF16
- Why this exists
- Provenance
- Quantization
- Measured performance
- Serving (vLLM XPU)
- Three things that will bite you
- Limitations — please read
- Licence
- 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
greglechin/Swift-1.5-Qwen3.8-27B-Uncensored-GPTQ-Int4-sym-G128-MTP-BF16 is a image-text-to-text model repository published by greglechin on Hugging Face. The source reports the transformers 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
- transformers
- Recent downloads (30 days)
- 0
- Cumulative likes
- 1
- Hugging Face trending score
- 1
- Reported safetensors parameters
- 27,781,427,952
- Architecture
- Qwen3_5ForConditionalGeneration
- License
- other
- Access
- Not gated by Hugging Face
- Created
- 2026-09-25T05:54:55.000Z
- Last modified
- 2026-09-25T05:59:27.000Z
Compatibility and lineage
Reported languages: en, zh
Reported base models: ajgazin/Swift-1.5-Qwen3.8-27B-Uncensored-MTP
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 2026-09-25T06:37:40.528Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.