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.
Qwen3.5-397B-A17B-MXFP4-AttnFP8-V2Florence-2-large-no-flash-attnMiniMax-M3-MXFP4-AttnFP8Kimi-K3-Quark-MXFP4-AttnFP8GLM-5.3-Flash-NVFP4-attn-lmhead-W4A16Qwen3.8-Flash-Next-NVFP4-W4A16-ATTN-FP8-MTP-NVFP4MiniMax-M3-MXFP4-Mixed-MXFP8-Attn-CT-AutoRoundgemma-4-E4B-it-Heretic-ARA-Refusals8-attngemma-3-4b-pt-obj-det-loc-tok-1024-single-stage-mm-proj-tokemb-attn-mlpllama3.2_3B_vl_kl0.5_mode1_attn_coef0.1_lr3e-4llama3.2_3B_vl_kl0.5_proj_attn_mode4_coef20.0_lr1e-4llama3.2_3B_vl_kl0.5_proj64_attn_mode4_coef5.0_lr1e-4llama3.2_3B_vl_kl0.5_mode1_attn_coef20.0_lr3e-4Qwen2-VL-7B-Instruct-bnb-4bit-r8-Attn-MLPllama3.2_3B_vl_kl0.5_mode1_attn_coef0.5_lr3e-4llama3.2_3B_vl_kl0.5_mode1_attn_coef5.0_lr3e-4llama3.2_3B_vl_kl0.5_proj64_attn_mode1_coef5.0_lr1e-4Qwen2_5vl_3B_AOP_train_attn_pooling_new_ablation1_1229_step5000gemma-4-E4B-it-Heretic-ARA-Refusals6-attnllama3.2_3B_vl_kl0.5_proj_attn_mode1_coef20.0_lr1e-4llama3.2_3B_vl_kl0.5_proj64_attn_mode1_coef0.5_lr1e-4lab_safety_qwen_3_5_2b_gspo_attn_guided_b_2_ga_4_ng_8lab_safety_qwen_3_5_point8b_gspo_attn_guided_b_2_ga_4_ng_8lab_safety_qwen_3_5_point8b_gspo_attn_guided_b_2_ga_4_ng_8_e_2Qwen2_5vl_3B_AOP_train_attn_pooling_new_ablation2_1229_step5000Swift-1.5-Qwen3.8-Flash-Next-W4A16-Attn8-FP8PLE
