scribekey/sotto-cleanup-350m-gguf
text-generation model by scribekey
Older snapshot: check the original source for current information.
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.
GGUF exports of juanquivilla/sotto-cleanup-lfm25-350m at revision 6df6f019170b8b55333c047b901886a51750a965, for on-device transcript cleanup with llama.cpp. ScribeKey uses the Q80 file for Smart Cleanup on Android. The weights are unchanged apart from GGUF conversion and quantisation. Sotto is a fine-tune of LiquidAI/LFM2.5-350M-Base by juanquivilla; see the upstream card for its training and evaluation. The model was trained with the tokenizer's BOS (). Either tokenize the string above with special-token parsing and no automatic BOS, or drop from the text and let llama.cpp add the BOS. Without a BOS the model answers "." for most inputs. Decode greedily and stop at EOS () or at ###; the cleaned text is everything before ###, trimmed. With converthftogguf.py from the llama.cpp tree vendored in llama-cpp-python 0.3.36: On ScribeKey's 149-sentence English dictation set, run exactly as the app runs it (prompt above, greedy, 256-token cap), Q80 scored the same as BF16 on…
Read the full model card ↗ · Preview checked 2026-10-28T22:41:45.679Z
Inside the original model card — Document outline
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
scribekey/sotto-cleanup-350m-gguf is a text-generation model repository published by scribekey 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
- text-generation
- Library
- gguf
- Recent downloads (30 days)
- 0
- Cumulative likes
- 0
- Hugging Face trending score
- 0
- Reported safetensors parameters
- Not reported
- Architecture
- Not reported
- License
- other
- Access
- Not gated by Hugging Face
- Created
- 2026-10-09T06:26:55.000Z
- Last modified
- 2026-10-09T06:27:11.000Z
Compatibility and lineage
Reported languages: en
Reported base models: juanquivilla/sotto-cleanup-lfm25-350m
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 2026-10-09T06:40:49.674Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.