File tree Expand file tree Collapse file tree
Expand file tree Collapse file tree Original file line number Diff line number Diff line change 7171||| qwen3_5_397b_a17b| ✅([ example] ( https://github.com/baidu-baige/LoongForge/tree/master/examples/qwen3.5/pretrain/pretrain_qwen3_5_397b_a17b.sh ) )| ✅([ example] ( https://github.com/baidu-baige/LoongForge/tree/master/examples/qwen3.5/finetuning/sft_qwen3_5_397b_a17b.sh ) )| [ model config] ( https://github.com/baidu-baige/LoongForge/tree/master/configs/models/qwen3.5/qwen3_5_397b_a17b.yaml ) |
7272|| Qwen3.6| qwen3_6_27b| ✅([ example] ( https://github.com/baidu-baige/LoongForge/tree/master/examples/qwen3.6/pretrain/pretrain_qwen3_6_27b.sh ) )| ✅([ example] ( https://github.com/baidu-baige/LoongForge/tree/master/examples/qwen3.6/finetuning/sft_qwen3_6_27b.sh ) )| [ model config] ( https://github.com/baidu-baige/LoongForge/tree/master/configs/models/qwen3.6/qwen3_6_27b.yaml ) |
7373||| qwen3_6_35b_a3b| ✅([ example] ( https://github.com/baidu-baige/LoongForge/tree/master/examples/qwen3.6/pretrain/pretrain_qwen3_6_35b_a3b.sh ) )| ✅([ example] ( https://github.com/baidu-baige/LoongForge/tree/master/examples/qwen3.6/finetuning/sft_qwen3_6_35b_a3b.sh ) )| [ model config] ( https://github.com/baidu-baige/LoongForge/tree/master/configs/models/qwen3.6/qwen3_6_35b_a3b.yaml ) |
74+ || Kimi-K2.5| kimi_k2_5| ✅([ example] ( https://github.com/baidu-baige/LoongForge/blob/master/examples/kimi_k2.5/pretrain/pretrain_kimi_k2.5_fp8.sh ) )| ✅([ example] ( https://github.com/baidu-baige/LoongForge/blob/master/examples/kimi_k2.5/sft/sft_kimi_k2.5_fp8.sh ) )| [ model config] ( https://github.com/baidu-baige/LoongForge/blob/master/configs/models/kimi_k2.5/kimi_k2_5.yaml ) |
7475|| LLaVA-OneVision-1.5| llava_onevision_1.5_4b|| ✅([ example] ( https://github.com/baidu-baige/LoongForge/tree/master/examples/llava_onevision/pretrain/stage_1_alignment_llava_ov_4b.sh ) )| [ model config] ( https://github.com/baidu-baige/LoongForge/tree/master/configs/models/llava_onevision/llava_onevision_1_5_4b.yaml ) |
7576|| InternVL-2.5| internvl2.5_8b|| ✅([ example] ( https://github.com/baidu-baige/LoongForge/tree/master/examples/internvl2.5/sft_internvl2_5_8b.sh ) )| [ model config] ( https://github.com/baidu-baige/LoongForge/tree/master/configs/models/internvl2.5/internvl2_5_8b.yaml ) |
7677||| internvl2.5_26b|| ✅([ example] ( https://github.com/baidu-baige/LoongForge/tree/master/examples/internvl2.5/sft_internvl2_5_26b.sh ) )| [ model config] ( https://github.com/baidu-baige/LoongForge/tree/master/configs/models/internvl2.5/internvl2_5_26b.yaml ) |
8485|| ERNIE4.5-VL| ernie4.5_vl_28b_a3b|| ✅([ example] ( https://github.com/baidu-baige/LoongForge/tree/master/examples/ernie4.5/finetuning/sft_ernie4.5_vl_28b_a3b.sh ) )| [ model config] ( https://github.com/baidu-baige/LoongForge/tree/master/configs/models/ernie4_5_vl/ernie4_5_vl_28b_a3b.yaml ) |
8586| Video| Wan2.2| wan2.2_i2v_a14b|| ✅([ example] ( https://github.com/baidu-baige/LoongForge/tree/master/examples/wan/pretrain_wan2.2_i2v_a14b.sh ) )| [ model config] ( https://github.com/baidu-baige/LoongForge/tree/master/configs/models/wan/wan2_2_i2v.yaml ) |
8687| VLA| Pi05| pi05|| ✅([ example] ( https://github.com/baidu-baige/LoongForge/tree/master/examples/pi05/finetuning/sft_pi05.sh ) )| [ model config] ( https://github.com/baidu-baige/LoongForge/tree/master/configs/models/pi05/pi05.yaml ) |
88+ || GR00T-N1.6| groot_n1_6|| ✅([ example] ( https://github.com/baidu-baige/LoongForge/tree/master/examples/groot_n1.6/finetuning/sft_groot.sh ) )| [ model config] ( https://github.com/baidu-baige/LoongForge/tree/master/configs/models/groot/groot_n1_6.yaml ) |
Original file line number Diff line number Diff line change @@ -51,3 +51,4 @@ LoongForge 支持在昆仑芯 P800 XPU 上进行训练,涵盖 LLM、VLM、VLA
5151| | | internvl3.5_241b_a28b | | ✅ (示例) |
5252| | Qwen-3.5 | qwen3.5_35b_a3b | ✅ (示例) | ✅ (示例) |
5353| VLA | PI 0.5 | | | ✅ (示例) |
54+ | | GR00T-N1.6 | | | ✅ (示例) |
You can’t perform that action at this time.
0 commit comments