CtrlK
BlogDocsLog inGet started
Tessl Logo

LLaVA-OneVision-2

github.com/EvolvingLMMs-Lab/LLaVA-OneVision-2

SkillAddedReview
commit-message

.opencode/skills/commit-message/SKILL.md

Guide for writing clear, consistent git commit messages following this repository's conventions

57

cu-lengths-attention-flow

.opencode/skills/cu-lengths-attention-flow/SKILL.md

Bilingual guide for understanding how cu_lengths controls attention behavior across ViT and LLM stages, and how patch_positions scope differs between the two

59

distributed-offline-packing

.opencode/skills/distributed-offline-packing/SKILL.md

Bilingual guide for running offline_packing/auto_pipe.sh across multiple nodes to produce padding-free packed WebDataset shards for SFT, with Energon Metadataset assembly

59

length-pool-sort-dataset

.opencode/skills/length-pool-sort-dataset/SKILL.md

Bilingual guide for understanding LengthPoolSortDataset cross-rank length synchronization mechanism in multi-GPU training

52

llava-onevision2-consistency

.opencode/skills/llava-onevision2-consistency/SKILL.md

Bilingual guide for running and interpreting LLaVA-OneVision2 HF vs Megatron consistency checks across TP and PP settings

60

megatron-checkpoint-layout

.opencode/skills/megatron-checkpoint-layout/SKILL.md

Bilingual guidance for Megatron checkpoint 1D 2D 3D mp_rank layouts across tensor pipeline and expert parallel dimensions

57

merge-ov2

.opencode/skills/merge-ov2/SKILL.md

Bilingual guide for merging ViT + LLM into LlavaOnevision2 HF checkpoint and validating weight/inference consistency

58

offline-packing-env-vars

.opencode/skills/offline-packing-env-vars/SKILL.md

Bilingual guide for the OFFLINE_PACKING_BMR and OFFLINE_PACKED_DATA environment variables that control LLaVA-OneVision2 training-side packing — what each gate does, why both must be enabled together, MBS=1 requirement, and the dead OFFLINE_PACKING_VQA branch

61