google/gemma-3-4b-it
Full fine-tune vs google/gemma-3-4b-pt. Fine-tune: 90% of the weights changed, across 34 of 34 layers.
Tensors changed: Embeddings 1/4, Norms 206/316, MLP 102/210, Attention 136/352, other 1/1
164 derivatives checked
- MaziyarPanahi/gemma-3-4b-it-GGUF: Same weights, converted (fewer layers)
- unsloth/gemma-3-4b-it-GGUF: Same weights, converted (fewer layers)
- unsloth/gemma-3-4b-it: Identical copy
- lmstudio-community/gemma-3-4b-it-GGUF: Quantization (fewer layers)
- ggml-org/gemma-3-4b-it-GGUF: Same weights, converted (fewer layers)
- bartowski/google_gemma-3-4b-it-GGUF: Same weights, converted (fewer layers)
- unsloth/gemma-3-4b-it-unsloth-bnb-4bit: Quantization
- nvidia/Nemotron-3-Content-Safety: LoRA-style
- sarvamai/sarvam-translate: Full fine-tune
- unsloth/gemma-3-4b-it-bnb-4bit: Quantization
- RedHatAI/gemma-3-4b-it-quantized.w4a16: Quantization
- google/gemma-3-4b-it-qat-q4_0-gguf: Full fine-tune (fewer layers)
- DavidAU/Gemma-3-4b-it-MAX-HORROR-Imatrix-GGUF: Quantization (fewer layers)
- nvidia/Nemotron-Content-Safety-Reasoning-4B: Full fine-tune
- gaianet/gemma-3-4b-it-GGUF: Same weights, converted (fewer layers)
- RedHatAI/gemma-3-4b-it-quantized.w8a8: Quantization
- mlabonne/gemma-3-4b-it-abliterated-GGUF: Abliterated (fewer layers)
- Mungert/X-Ray_Alpha-GGUF: Full fine-tune (fewer layers)
- Indexnusrefather/gemma-3-4b-it-roleplay-tuned-v2: LoRA-style
- Aldaris/gemma-3-4b-it-Q4_K_M-GGUF: Quantization (fewer layers)
- lmstudio-community/gemma-3-4B-it-qat-GGUF: Full fine-tune (fewer layers)
- DavidAU/Gemma-3-4b-it-MAX-NEO-Imatrix-GGUF: Quantization (fewer layers)
- mradermacher/gemma-3-4b-it-GGUF: Same weights, converted (fewer layers)
- matrixportalx/gemma-3-4b-it-abliterated-GGUF: Abliterated (fewer layers)
- mradermacher/gemma-3-4b-it-i1-GGUF: Quantization (fewer layers)
- Mungert/sarvam-translate-GGUF: Full fine-tune (fewer layers)
- AdaptLLM/biomed-gemma-3-4b-it: Full fine-tune
- zw89/gemma-3-4b-it-GGUF: Same weights, converted (fewer layers)
- saidutta69/gemma-3-4b-it-heretic: Abliterated
- INSAIT-Institute/MamayLM-Gemma-3-4B-IT-v1.0: Full fine-tune
67 other repos hold byte-for-byte identical weight files.