LiquidAI/LFM2.5-1.2B-Instruct
Full fine-tune vs LiquidAI/LFM2.5-1.2B-Base. 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
100 derivatives checked
- LiquidAI/LFM2.5-1.2B-Instruct-GGUF: Same weights, converted
- lmstudio-community/LFM2.5-1.2B-Instruct-MLX-8bit: Quantization
- lmstudio-community/LFM2.5-1.2B-Instruct-MLX-4bit: Quantization
- lmstudio-community/LFM2.5-1.2B-Instruct-MLX-6bit: Quantization
- unsloth/LFM2.5-1.2B-Instruct-GGUF: Same weights, converted
- lmstudio-community/LFM2.5-1.2B-Instruct-GGUF: Quantization
- unsloth/LFM2.5-1.2B-Instruct: Identical copy
- mlx-community/LFM2.5-1.2B-Instruct-4bit: Quantization
- NovachronoAI/LFM2.5-1.2B-Nova-Function-Calling-GGUF: Quantization
- bartowski/LiquidAI_LFM2.5-1.2B-Instruct-GGUF: Same weights, converted
- LiquidAI/LFM2.5-1.2B-Instruct-MLX-4bit: Quantization
- d4rkninja/tanpo-marketing: Partial fine-tune
- LiquidAI/LFM2.5-1.2B-Instruct-MLX-8bit: Quantization
- d4rkninja/tanpo-product: Partial fine-tune
- d4rkninja/tanpo-marketing-GGUF: LoRA-style
- NobodyWho/LFM2.5-1.2B-Instruct-GGUF: Same weights, converted
- d4rkninja/tanpo-deals: LoRA-style
- d4rkninja/tanpo-hiring: Partial fine-tune
- LiquidAI/LFM2.5-1.2B-Instruct-DSpark: Different architecture
- ermiaazarkhalili/LFM2.5-1.2B-SFT-Claude-Opus-Reasoning-Unsloth-GGUF: Full fine-tune
- d4rkninja/tanpo-fundraising: LoRA-style
- mradermacher/LFM2.5-1.2B-Instruct-i1-GGUF: Quantization
- alpharomercoma/LFM2.5-1.2B-Instruct-heretic: Abliterated
- zaakirio/LFM2.5-1.2B-Instruct-Uncensored: Abliterated
- d4rkninja/tanpo-ops: LoRA-style
- Indexnusrefather/Palette-RP-1.2B-Instruct-2609-v0.5: Full fine-tune
- Firworks/LFM2.5-1.2B-Instruct-nvfp4: Quantization
- richardyoung/LFM2.5-1.2B-Instruct-heretic: Abliterated
- d4rkninja/tanpo-finance: LoRA-style
- Indexnusrefather/Super-Slop-Machina-Roleplay-1.2b: LoRA-style
9 other repos hold byte-for-byte identical weight files.