Qwen3vl 4b Int8 Convrot
Convrot conversion of qwen3vl 4b. near identical to bf16 in terms of adherence and quality and usually near identical output to bf16. the learned version might slightly outperform the simple. but simple handles the...
Model source
- Kind
- Text model
- File size
- 4.89 GB
- Base model
- Qwen/Qwen3-VL-4B-Instruct
- Version
- vINT8
- Updated
- Oct 18, 2026
- Source
- Hugging Face
Source description
Convrot conversion of qwen3vl 4b. near identical to bf16 in terms of adherence and quality and usually near identical output to bf16. the learned version might slightly outperform the simple. but simple handles the bf16 layers differently and you might prefer that one.
optimized fork of Ai-Toolkit
https://github.com/PlagueKind/BlackTrainer
If you want to support my work I accept XMR monero
865BrcfWLdwELwuq5faV1uVTbh93zVK6AUYLY2c3mX6sFfAGRfS6axe1kBTYYKuM7ccN7zBZDAZvnT7E4NKmUazySdbpc7p
Sources
1 source
Sources
1 sourceVerified Oct 18
Model artifacts
1 artifact
Model artifacts
1 artifactqwen3vl_4b_int8_convrot_learned.safetensors
safetensors · INT8 · 4.89 GB · SHA-256 6fe6c6aa2f1b…f534 · Hugging Face
DownloadSource excerpts
2 excerpts
Source excerpts
2 excerpts--- base_model: Qwen/Qwen3-VL-4B-Instruct tags: - qwenvl - qwen - text - encoder - krea - krea2 --- Convrot conversion of qwen3vl 4b. near identical to bf16 in terms of adherence and quality and usually near identical output to bf16. the learned version might slightly outperform the simple. but simple handles the bf16 layers differently and you might prefer that one. optimized fork of Ai-Toolkit https://github.com/PlagueKind/BlackTrainer If you want to support my work I accept XMR monero ```865BrcfWLdwELwuq5faV1uVTbh93zVK6AUYLY2c3mX6sFfAGRfS6axe1kBTYYKuM7ccN7zBZDAZvnT7E4NKmUazySdbpc7p```
Source context: 0 downloads · 4 likes · Repo Plaguekind/qwen3vl-4b-int8-convrot
