tiiuae/Falcon3-7B-Instruct
Full fine-tune vs tiiuae/Falcon3-7B-Base. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 57/57, MLP 84/84, Attention 112/112, lm_head 1/1
30 derivatives checked
- bartowski/Falcon3-7B-Instruct-GGUF: Same weights, converted
- tiiuae/Falcon3-7B-Instruct-1.58bit: Quantization
- mradermacher/Falcon3-7B-Instruct-i1-GGUF: Quantization
- tiiuae/Falcon3-7B-Instruct-GGUF: Same weights, converted
- second-state/Falcon3-7B-Instruct-GGUF: Same weights, converted
- ChiKoi7/Falcon3-7B-Instruct-Heretic-GGUF: Abliterated
- ChiKoi7/Falcon3-7B-Instruct-Heretic-v2-GGUF: Abliterated
- mradermacher/Falcon3-7B-Instruct-GGUF: Same weights, converted
- mlx-community/Falcon3-7B-Instruct-4bit: Quantization
- gaianet/Falcon3-7B-Instruct-GGUF: Same weights, converted
- s3dev-ai/Falcon3-7B-Instruct-gguf: Same weights, converted
- lmstudio-community/Falcon3-7B-Instruct-GGUF: Quantization
- EasierAI/Falcon-3-7B: Quantization
- ehristoforu/Falcon3-MoE-2x7B-Insruct: MoE built from dense models
- RedaAlami/Falcon3-7B-Instruct-Distill-DS-v1: Full fine-tune
- DevQuasar-6/tiiuae.Falcon3-7B-Instruct-GGUF: Quantization
- tiiuae/Falcon3-7B-Instruct-GPTQ-Int8: Quantization
- tiiuae/Falcon3-7B-Instruct-AWQ: Quantization
- tiiuae/Falcon3-7B-Instruct-GPTQ-Int4: Quantization
- neopolita/jessi-v0.3-falcon3-7b-instruct: Full fine-tune
- ChiKoi7/Falcon3-7B-Instruct-Heretic: Abliterated
- ChiKoi7/Falcon3-7B-Instruct-Heretic-v2: Abliterated
- neopolita/jessi-v0.6-falcon3-7b-instruct: Full fine-tune
- ehristoforu/Phi4-MoE-2x14B-Instruct: Different architecture
- neopolita/jessi-v0.5-falcon3-7b-instruct: Full fine-tune
- neopolita/jessi-v0.4-falcon3-7b-instruct: Full fine-tune
- neopolita/jessi-v0.2-falcon3-7b-instruct: Full fine-tune
- jahyungu/Falcon3-7B-Instruct_mathqa: Full fine-tune
- jahyungu/Falcon3-7B-Instruct_openbookqa: Full fine-tune
- MaziyarPanahi/falcon3-7b-reasoning-v0.1: LoRA-style
1 other repos hold byte-for-byte identical weight files.