coolthor/VoxCPM2-Q4_K_M-GGUF
text-to-speech model by coolthor
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.
The main file is a mixed 4-bit quantization: a Q4KS base with attention V and FFN down-projection weights kept at Q6K in all 28 layers. It replaces the previous VoxCPM2-BaseLM-Q4KM.gguf release. We encountered occasional whole-sentence breakdowns with the previous Q4KM in production, especially on long sentences: the voice became distant and the latter part became unintelligible. We withdrew that version from production. Its earlier short listening test did not catch this problem. The protected recipe is the smallest of the three replacements evaluated here; the results below support choosing it, but do not prove that rare failures have been eliminated. These files contain the BaseLM for llama.cpp-omni (tools/omni/voxcpm2). The repository name retains Q4KM for continuity; the protected file uses the different recipe described above. You also need VoxCPM2-Acoustic-F16.gguf from DennisHuang648/VoxCPM2-GGUF. The acoustic stack stays at F16 and is not re-hosted here. Sizes above are for BaseLM only. SHA-256: Each recipe…
Read the full model card ↗ · Preview checked 2026-10-28T00:18:51.660Z
Inside the original model card — Document outline
- VoxCPM2 BaseLM — protected 4-bit GGUF for llama.cpp-omni
- Files
- Evaluation (RTX 5090, sm120)
- Usage
- Quantization details
- License and attribution
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: apache-2.0. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
Model overview
coolthor/VoxCPM2-Q4_K_M-GGUF is a text-to-speech model repository published by coolthor 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-to-speech
- Library
- gguf
- Recent downloads (30 days)
- 0
- Cumulative likes
- 0
- Hugging Face trending score
- 0
- Reported safetensors parameters
- Not reported
- Architecture
- Not reported
- License
- apache-2.0
- Access
- Not gated by Hugging Face
- Created
- 2026-10-19T06:17:48.000Z
- Last modified
- 2026-10-19T06:18:29.000Z
Compatibility and lineage
Reported languages: zh, en
Reported base models: openbmb/VoxCPM2
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 2026-10-19T06:43:13.200Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.