agentmods.dev

LLaVA-OneVision-2: Skill for OpenCode

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

llava-onevision2-consistency is a skill for OpenCode from EvolvingLMMs-Lab/LLaVA-OneVision-2. It costs 32 tokens per session (5,119 once invoked), scanned A, original, Apache-2.0.

A bilingual guide for checking whether two versions of a LLaVA-OneVision2 AI model behave consistently: one using Hugging Face and one using Megatron/MCore. It covers tests across tensor and pipeline parallel settings, which split model work across hardware.

In plain words
What is it for?
Running the repository's recommended tests, converting checkpoints when needed, and comparing Hugging Face and Megatron/MCore model behavior across different parallel configurations.
Why use it?
Converting a model between these systems can change its behavior without making the problem obvious. Consistency checks help detect such differences before relying on the converted model.

Skill for OpenCode

Written for OpenCode: installed under .opencode/. Also seen: mentions OpenCode.

This is EvolvingLMMs-Lab/LLaVA-OneVision-2's own configuration. It tells OpenCode how to work on LLaVA-OneVision-2 itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything LLaVA-OneVision-2 configures →

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is bash tests/consistency/run_consistency_tests.sh.

not rated 1.2krepo +5 yesterday A scan SkillSpector: pass 32 tokens original Apache-2.0
View source ↗ EvolvingLMMs-Lab/LLaVA-OneVision-2 evolvinglmms-lab.github.io
About the project

LLaVA-OneVision-2 is an openly released multimodal AI model and training framework that processes images, long-form video, and spatial information. Researchers use it to train, evaluate, and reproduce vision-language models with the project’s released data, encoders, checkpoints, and training records. The catalogue skills support work with this model and its training resources.

EvolvingLMMs-Lab/LLaVA-OneVision-2 · 1,232 stars · on GitHub · evolvinglmms-lab.github.io

Reuse

Borrowing it

Nothing to install: this file belongs to EvolvingLMMs-Lab/LLaVA-OneVision-2. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/EvolvingLMMs-Lab/LLaVA-OneVision-2/main/.opencode/skills/llava-onevision2-consistency/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/EvolvingLMMs-Lab/LLaVA-OneVision-2

Made for: OpenCode.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for llava-onevision2-consistency

README.md
[![agentmods](https://agentmods.dev/badge/skills/evolvinglmms-lab/llava-onevision-2/llava-onevision2-consistency/github.svg)](https://agentmods.dev/skills/evolvinglmms-lab/llava-onevision-2/llava-onevision2-consistency)
Your own site
<a href="https://agentmods.dev/skills/evolvinglmms-lab/llava-onevision-2/llava-onevision2-consistency"><img src="https://agentmods.dev/badge/skills/evolvinglmms-lab/llava-onevision-2/llava-onevision2-consistency/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for llava-onevision2-consistency

Your own site · 80×15
<a href="https://agentmods.dev/skills/evolvinglmms-lab/llava-onevision-2/llava-onevision2-consistency"><img src="https://agentmods.dev/badge/skills/evolvinglmms-lab/llava-onevision-2/llava-onevision2-consistency.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 32 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 5,119 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00032 $0.05119
Opus 5.5 $0.00013 $0.02048
Sonnet 5.5 $0.00006 $0.01024
Haiku 4.5 $0.00003 $0.00512

Measured 2mo ago against content hash f4767c5faae1, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

llava-onevision2-consistency scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.opencode/skills/llava-onevision2-consistency/SKILL.md · 372 lines

How it starts

The opening of the file, as written. The whole thing — 372 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Purpose / 用途

Use this skill when validating whether a HuggingFace checkpoint and a Megatron/MCore checkpoint are behaviorally consistent in this repository.

在这个仓库里，需要验证 HuggingFace checkpoint 和 Megatron/MCore checkpoint 是否行为一致时，使用这个 skill。

There are two test systems in this repo:

本仓库有两套测试系统：

1. pytest test suite (recommended / 推荐)

  • tests/consistency/conftest.py — session fixtures, HF→mcore conversion, Megatron initialization

  • tests/consistency/test_model_consistency.py — 6 integration tests

  • tests/consistency/test_consistency_utils.py — 10 utility functions + 11 unit tests

  • tests/consistency/run_consistency_tests.sh — shell wrapper with auto-conversion + torchrun

  • tests/consistency/conftest.py —— session 级 fixture、HF→mcore 转换、Megatron 初始化

  • tests/consistency/test_model_consistency.py —— 6 个集成测试

  • tests/consistency/test_consistency_utils.py —— 10 个工具函数 + 11 个单元测试

  • tests/consistency/run_consistency_tests.sh —— shell 入口，自动转换 + torchrun

2. Legacy monolithic script (reference only / 仅供参考)

  • examples/llava_onevision2/check_model_consistency.sh
  • examples/llava_onevision2/check_model_consistency.py

仅作历史参考，新的工作请用 pytest 套件。

Architecture / 架构

Direction: HF → mcore

The pytest suite assumes only the HF checkpoint exists as input. The mcore checkpoint is generated automatically via conversion.

pytest 测试套件假设 只有 HF checkpoint 作为输入。mcore checkpoint 通过转换 自动生成。

HF auto-model (input)
  → convert_4b_hf_to_mcore.sh (auto-run by conftest.py or run_consistency_tests.sh)
    → mcore checkpoint (generated)
      → both models loaded → 6 tests run

Direction: mcore → HF (reverse / deploy / round-trip) / 反向：mcore → HF（部署 / 回环）

The pytest suite does not exercise the reverse path. For the p14m2 variant, two scripts ship for this:

Script Use case
examples/llava_onevision2/convert/convert_4b_p14m2_mcore_to_hf.sh Single mcore→HF pass (deploy, inference debug)
examples/llava_onevision2/convert/convert_4b_p14m2_mcore_to_release.sh Re-shard mcore via HF round-trip (change TP/PP without retraining)

Read the full file on GitHub · 372 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2mo ago First seen · 372 lines · 32 tokens per session scan A f4767c5faae1

Subscribe to this mod's changes

llava-onevision2-consistency is a skill published in the GitHub repository EvolvingLMMs-Lab/LLaVA-OneVision-2 (1,232 stars, last pushed yesterday), licensed Apache-2.0. It adds 32 tokens to every session and 5,119 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Data and AILLMTestingllavallava-onevisionmllmqwen3vision-language-model
Related

llava-onevision2-consistency compared with skills from other repositories

llava-onevision2-consistency next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
llava-onevision2-consistencyEvolvingLMMs-Lab/LLaVA-OneVision-2Skill1.2kyesterdayApache-2.0A32OpenCode
add-new-modelBlaizzy/mlx-vlmSkill5.7k4d agoMITA83Claude Code
cli-inferenceBlaizzy/mlx-vlmSkill5.7k4d agoMITA67Claude Code
convert-quantizeBlaizzy/mlx-vlmSkill5.7k4d agoMITA99Claude Code
server-inferenceBlaizzy/mlx-vlmSkill5.7k4d agoMITA80Claude Code
hf-cache-modelsBlaizzy/mlx-vlmSkill5.7k4d agoMITA38Claude Code
mle-workflowaffaan-m/ECCSkill298k4d agoMITA44Claude Code, Codex