google/gemma-3-27b-it
Full fine-tune vs google/gemma-3-27b-pt. Fine-tune: 98% of the weights changed, across 62 of 62 layers.
Tensors changed: Embeddings 1/4, Norms 374/484, MLP 186/294, Attention 248/464, other 1/1
92 derivatives checked
- gaunernst/gemma-3-27b-it-int4-awq: Quantization
- ISTA-DASLab/gemma-3-27b-it-GPTQ-4b-128g: Quantization
- RedHatAI/gemma-3-27b-it-quantized.w4a16: Quantization
- MaziyarPanahi/gemma-3-27b-it-GGUF: Quantization (fewer layers)
- RedHatAI/gemma-3-27b-it-FP8-dynamic: Quantization
- unsloth/gemma-3-27b-it: Identical copy
- unsloth/gemma-3-27b-it-GGUF: Same weights, converted (fewer layers)
- bartowski/google_gemma-3-27b-it-GGUF: Same weights, converted (fewer layers)
- abhishekchohan/gemma-3-27b-it-quantized-W4A16: Quantization
- leon-se/gemma-3-27b-it-FP8-Dynamic: Quantization
- mlabonne/gemma-3-27b-it-abliterated-GGUF: Full fine-tune (fewer layers)
- m8than/gemma-3-27b-lenientchatfix: Identical copy
- mlabonne/gemma-3-27b-it-abliterated: Abliterated
- lmstudio-community/gemma-3-27b-it-GGUF: Quantization (fewer layers)
- NbAiLab/borealis-27b: Full fine-tune
- ggml-org/gemma-3-27b-it-GGUF: Same weights, converted (fewer layers)
- unsloth/gemma-3-27b-it-bnb-4bit: Quantization
- gaunernst/gemma-3-27b-it-qat-compressed-tensors: Quantization
- lmstudio-community/gemma-3-27B-it-qat-GGUF: Full fine-tune (fewer layers)
- unsloth/gemma-3-27b-it-unsloth-bnb-4bit: Quantization
- TheDrummer/Big-Tiger-Gemma-27B-v3-GGUF: Quantization (fewer layers)
- Alevnokc/Gemma-3-27B-Roblox-Luau: Full fine-tune (fewer layers)
- TheDrummer/Gemma-3-R1-27B-v1-GGUF: Quantization (fewer layers)
- mradermacher/gemma-3-27b-it-i1-GGUF: Quantization (fewer layers)
- chatpig/gemma-3-27b-it-gguf: Quantization (fewer layers)
- qwp4w3hyb/gemma-3-27b-it-iMat-GGUF: Same weights, converted (fewer layers)
- Thaurock/Gemma-3-27b-it-abliterated-GGUF: Abliterated (fewer layers)
- second-state/gemma-3-27b-it-GGUF: Same weights, converted (fewer layers)
- Mungert/gemma-3-27b-it-qat-q4_0-GGUF: Full fine-tune (fewer layers)
- tripolskypetr/gemma-3-27b-it: Same weights, converted (fewer layers)
17 other repos hold byte-for-byte identical weight files.