models
Open weights, fine-tunes and adapters. Every listing here comes live from the Hugging Face Hub, attributed to it, and links back to the source.
Llama-3.1-Minitron-4B-Depth-Baseslimder-qwen38-reap384-depth40-mb4-5slimder-qwen38-ream288-depth32-agentic-ngram50-compact-nvfp4OLMo3-3B-SiameseNorm-DepthAttention-stage2OLMo3-3B-SiameseNorm-DepthAttention-stage1OLMo3-3B-SiameseNorm-DepthAttention-stage4-instructOLMo3-3B-SiameseNorm-DepthAttention-stage4-thinkOLMo3-3B-SiameseNorm-DepthAttention-stage3Llama-3.1-Minitron-4B-Depth-BIA-proof-of-concepthubble-1b-100b_toks-double_depth-perturbed-hfhubble-1b-100b_toks-double_depth-standard-hfnvidia_Llama-3.1-Minitron-4B-Depth-Base-GGUFhubble-1b-100b_toks-half_depth-standard-hfQwen3-1.7B-Depth-Aggressive-Q4_K_M-GGUFhubble-1b-100b_toks-half_depth-perturbed-hfOuro-1.4B-Thinking-depth-SFTOuro-1.4B-Thinking-depth-GRPOlatent-recurrent-depth-lmLlama-3.1-Minitron-4B-Depth-Neo-BAAI-100kJetMoE_rank_lstm_full_trained_depth3_n207_vaswani_RoPE_hi_hf_frames_heads_4_layers_4_random_depth_3Llama-3.1-Minitron-4B-Depth-Base-AutoRound-GPTQ-sym-4bitLlama-3.1-Minitron-4B-Depth-Base-AutoRound-GPTQ-asym-4bitQwen3-4B-Instruct-2507-RLM-RLVR-FullFT-lr5e-6-depth1-v108_GPT2_RoPE_hi_hf_frames_heads_4_layers_4_random_depth_3SmolLM-360M-width-depth-pruned-to-80MJetMoE_rank_lstm_full_trained_depth3_n2_before_switchgiannisan_Mistral-10.7B-Instruct-v0.3-depth-upscaling-4_0bpw_exl2JetMoE_rank_lstm_final_full_trained_depth3_n2OLMo3-1B-SiameseNorm-DepthAttention-stage3