Instructions to use mlai-dante/road-model-public with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use mlai-dante/road-model-public with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-VL-8B-Instruct") model = PeftModel.from_pretrained(base_model, "mlai-dante/road-model-public") - Notebooks
- Google Colab
- Kaggle
Ctrl+K
- churro
- churro3b_alpha16_s42_beam5_20260805_200754_full
- churro3b_alpha2_s42_beam5_20260807_181928_full
- churro3b_anchor002_hires_alpha4_2ep_20260807_041937_full
- churro3b_colabnew_lf_aug_neftune_peft_2ep_20260801_022739_full
- churro3b_ctc_alignment_alpha4_2ep_v1
- churro3b_dora_2ep_20260801_174958_full
- churro3b_fullrows_r64_visualattn_r32_a64_2ep_20260730_030635_full
- churro3b_fullrows_r64_visualattn_r32_a64_2ep_20260730_213327_full
- churro3b_fullrows_r64_visualattn_r32_a64_2ep_20260731_173357_full
- churro3b_fullrows_r64_visualattn_r32_a64_4ep_20260731_012126_full
- churro3b_fullrows_r64_visualattn_r32_a64_4ep_20260731_043006_full
- churro3b_fullrows_r64_visualattn_r32_a64_visualmerger_r16_a32_2ep_20260730_164353_full
- churro3b_fullrows_r64_visualattn_r32_a64_visualmerger_r16_a32_2ep_20260730_195006_full
- churro3b_fullrows_r64_visualattn_r8_a16_2ep_20260729_215334_fold0
- churro3b_fullrows_r64_visualattn_r8_a16_2ep_20260729_215334_full
- churro3b_global4_hires_alpha4_s42_beam8_20260807_013428_full
- churro3b_globalrss_hires_alpha4_2ep_20260809_010059_full
- churro3b_hires_alpha4_plus1ep_20260803_130100_full
- churro3b_light_2ep_20260801_222718_full
- churro3b_marker_s42_beam5_20260805_161223_full
- churro3b_merger_unsloth_r64_visattn_r32_2ep_20260801_004606_full
- churro3b_modality_split_langr32_s1337_beam8_20260805_053625_full
- churro3b_page8_hires_alpha4_2ep_20260806_040046_full
- churro3b_pseudo6of8_2ep_20260801_151155_full
- churro3b_quality_drop_r128_a256_2ep_20260728_163759_fold0
- churro3b_quality_drop_r128_a256_2ep_20260728_163759_full
- churro3b_quality_drop_r64_a128_2ep_20260728_032905_fold0
- churro3b_quality_drop_r64_a128_2ep_20260728_032905_full
- churro3b_r64_multiview_20260729_031442
- churro3b_road_center_focus_drop_20260726_160600
- churro3b_road_original_fold0_20260716_141131_64d9a4
- churro3b_road_original_full_2epoch_standalone_20260716_204724_2d5155
- churro3b_road_original_full_4epoch_standalone_20260716_234456_1deb5e
- churro3b_swa5_hires_alpha4_2ep_20260804_044441_full
- churro3b_sweep_alpha4_2ep_20260802_025242_full
- churro3b_sweep_hires_alpha8_2ep_20260802_232853_full
- churro3b_sweep_hires_dropout10_2ep_20260803_025306_full
- churro3b_sweep_hires_fullrows_2ep_20260803_190842_full
- churro3b_sweep_hires_lmhead_2ep_20260803_233700_full
- churro3b_sweep_hires_maxcap1024_2ep_20260804_024142_full
- churro3b_sweep_hires_quality_all_2ep_20260803_162438_full
- churro3b_sweep_hires_xl_alpha4_2ep_20260802_200637_full
- churro3b_tiedembed_r16_4ep_20260806_153511_full
- churro3b_vision_dora32_alllinear_min384_2ep_20260729_050754_fold0
- churro3b_xheight_cont2ep_lr5e5_20260808_030846_full
- churro3b_xheight_xheight_full_20260808_003645
- experiments
- gemma4_aug
- llava_onevision_7b_road_fold0