mistralai/Ministral-3-3B-Reasoning-2512
Full fine-tune vs mistralai/Ministral-3-3B-Base-2512. Fine-tune: 100% of the weights changed, across 26 of 26 layers.
Tensors changed: Embeddings 1/1, Norms 103/103, MLP 150/150, Attention 200/200, other 4/4
21 derivatives checked
- MaziyarPanahi/Ministral-3-3B-Reasoning-2512-GGUF: Same weights, converted (fewer layers)
- unsloth/Ministral-3-3B-Reasoning-2512-GGUF: Same weights, converted (fewer layers)
- mistralai/Ministral-3-3B-Reasoning-2512-GGUF: Same weights, converted (fewer layers)
- bartowski/mistralai_Ministral-3-3B-Reasoning-2512-GGUF: Same weights, converted (fewer layers)
- lmstudio-community/Ministral-3-3B-Reasoning-2512-GGUF: Quantization (fewer layers)
- ggml-org/Ministral-3-3B-Reasoning-2512-GGUF: Quantization (fewer layers)
- unsloth/Ministral-3-3B-Reasoning-2512-bnb-4bit: Quantization
- mradermacher/Ministral-3-3B-Reasoning-2512-i1-GGUF: Quantization (fewer layers)
- unsloth/Ministral-3-3B-Reasoning-2512-unsloth-bnb-4bit: Quantization
- EnlistedGhost/Ministral-3-3B-Reasoning-2512-GGUF: Same weights, converted (fewer layers)
- unsloth/Ministral-3-3B-Reasoning-2512: Identical copy
- jkim96/Ministral-3-3B-Reasoning-2512-DASHQ-Q2-GGUF: Quantization (fewer layers)
- mradermacher/Ministral-3-3B-Reasoning-2512-GGUF: Same weights, converted (fewer layers)
- s3dev-ai/Ministral-3-3B-Reasoning-2512-gguf: Same weights, converted (fewer layers)
- Jeney/Ministral-3-3B-Reasoning-2512-GGUF: Same weights, converted (fewer layers)
- DevQuasar/mistralai.Ministral-3-3B-Reasoning-2512-GGUF: Same weights, converted (fewer layers)
- cyankiwi/Ministral-3-3B-Reasoning-2512-AWQ-4bit: Quantization
- ertghiu256/ministral-3-3b-glm-distillation: LoRA-style
- ValiantLabs/Ministral-3-3B-Reasoning-2512-Esper3.1: LoRA-style
- MuXodious/Ministral-3-3B-Reasoning-2512-absolute-heresy: Abliterated
- coder3101/Ministral-3-3B-Reasoning-2512-heretic: Abliterated