JamyDohrn/LTE-Qwen3-4B-Base
Different architecture vs Qwen/Qwen3-8B-Base. Different architecture: only 72 of 399/399 tensors match by name and shape.
2 derivatives checked
- mradermacher/LTE-Qwen3-4B-Base-i1-GGUF: Quantization
- mradermacher/LTE-Qwen3-4B-Base-GGUF: Same weights, converted
1 other repos hold byte-for-byte identical weight files.