EvolvingLMMs-Lab

9 mods across 2 repositories, 1.2k stars between them.

EvolvingLMMs-Lab/LLaVA-OneVision-2

Skill Claude CodeCodex

Bilingual guide for understanding how culengths controls attention behavior across ViT and LLM stages, and how patchpositions scope differs between the two.

1.2k 2d ago A 35 tokens original Apache-2.0

EvolvingLMMs-Lab/LLaVA-OneVision-2

Skill Claude CodeCodex

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

1.2k 2d ago C 40 tokens original Apache-2.0

merge-ov2

07

EvolvingLMMs-Lab/LLaVA-OneVision-2

Skill Claude CodeCodex

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

1.2k 2d ago A 29 tokens original Apache-2.0

EvolvingLMMs-Lab/LLaVA-OneVision-2

Skill Claude CodeCodex

Bilingual guide for the OFFLINEPACKINGBMR and OFFLINEPACKEDDATA 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 OFFLINEPACKINGVQA branch.

1.2k 2d ago A 65 tokens original Apache-2.0

engram AGENTS.md

09

EvolvingLMMs-Lab/engram

Instructions file CodexOpenCode

AGENTS.md instructions for EvolvingLMMs-Lab/engram, covering agents.md, repository management, toolchain, common commands and dependencies.

23 2mo ago A 463 tokens