Hcompany/Holo3-35B-A3B
Full fine-tune (fewer layers) vs Qwen/Qwen3.5-35B-A3B. Hcompany/Holo3-35B-A3B matches part of Qwen/Qwen3.5-35B-A3B (1026 of its 1811 tensors; the rest of Qwen/Qwen3.5-35B-A3B has no counterpart). On the shared part: Fine-tune: 100% of the weights changed, across 40 of 40 layers.
Tensors changed: MLP 389/392, lm_head 1/1, Embeddings 4/4, Attention 367/388, Norms 207/241
9 derivatives checked
- mudler/Holo3-35B-A3B-APEX-GGUF: Quantization
- mradermacher/Holo3-35B-A3B-i1-GGUF: Quantization
- mradermacher/Holo3-35B-A3B-GGUF: Quantization
- deadbydawn101/RavenX-OpenFable-Qwopus-Coder-Holo3-Qwen3.6-35B-A3B-MTP: Full fine-tune
- nightmedia/Qwen3.6-35B-A3B-Holo3-Qwopus-Instruct-qx64-hi-mlx: Quantization
- nightmedia/Qwen3.6-35B-A3B-MTP-Holo3-Qwopus-Coder-qx64y-hi-mlx: Quantization
- nightmedia/Qwen3.6-35B-A3B-Qwable-Holo3-Qwopus: Full fine-tune
- DevQuasar/Hcompany.Holo3-35B-A3B-GGUF: Quantization
- deadbydawn101/RavenX-OpenFable-Qwopus-Coder-Holo3-Qwen3.6-35B-A3B-MTP-mlx: Quantization
1 other repos hold byte-for-byte identical weight files.