LiquidAI/LFM2.5-VL-1.6B
Full fine-tune (fewer layers) vs LiquidAI/LFM2.5-1.2B-Base. LiquidAI/LFM2.5-1.2B-Base matches part of LiquidAI/LFM2.5-VL-1.6B (148 of its 589 tensors; the rest of LiquidAI/LFM2.5-VL-1.6B has no counterpart). On the shared part: Fine-tune: 100% of the weights changed, across 16 of 16 layers.
Tensors changed: Embeddings 2/2, other 30/30, MLP 48/48, Norms 44/44, Attention 24/24
18 derivatives checked
- LiquidAI/LFM2.5-VL-1.6B-GGUF: Same weights, converted (fewer layers)
- ente-ai/LFM2.5-VL-1.6B-GGUF: Same weights, converted (fewer layers)
- unsloth/LFM2.5-VL-1.6B-GGUF: Same weights, converted (fewer layers)
- NobodyWho/LFM2.5-VL-1.6B-GGUF: Same weights, converted (fewer layers)
- lmstudio-community/LFM2.5-VL-1.6B-MLX-6bit: Quantization
- lmstudio-community/LFM2.5-VL-1.6B-GGUF: Quantization (fewer layers)
- DevQuasar/LiquidAI.LFM2.5-VL-1.6B-GGUF: Same weights, converted (fewer layers)
- LiquidAI/LFM2.5-VL-1.6B-Extract: Full fine-tune
- lewiswatson/Frame2KG-LFM-2.5-1.6b-JSON: Partial fine-tune (fewer layers)
- lmstudio-community/LFM2.5-VL-1.6B-MLX-8bit: Quantization
- lmstudio-community/LFM2.5-VL-1.6B-MLX-4bit: Quantization
- minhduc168/LFM2.5-VL-1.6B-Vietnamese-LoRA: Adapter
- minhduc168/LFM2.5-VL-1.6B-Vietnamese-bill-diagnosis-LoRA: Adapter
- Yonesuke0716/lfm-iqa-gguf: Quantization (fewer layers)
- RKNNAI/RK182X-VLM-LFM2.5-VL-1.6B: Quantization
- MuXodious/LFM2.5-VL-1.6B-absolute-heresy-MPOA: Abliterated
- MuXodious/LFM2.5-VL-1.6B-absolute-heresy: Abliterated
- coder3101/LFM2.5-VL-1.6B-heretic: Abliterated