A Models settings companion page listing every model's input modalities in one place, editable for both LLM adapter families

OpenCode plugin that enriches model entries with full metadata (context length, pricing, modalities, capability flags) fetched from a provider-supplied OpenRouter-shaped endpoint.

DSH plugin: auto-detect and configure model capabilities (reasoningEfforts + input modalities) for llm-pi-ai. Successor to dsh-reasoning-efforts.

Data-navigator is a JavaScript library that allows for serial navigation of data structures using a variety of input modalities and assistive technologies.

Model Capability Manager for DSH Web: per-model thinking levels, context window, output caps, input modalities, gateway compat, route defaults, one-click presets, EN/中文 UI.

Framework-free registry of exact AI model capabilities, reasoning wire params, modalities, and limits (Anthropic / OpenAI / Google + proxy routes).

Live & generative rich-media for @warlock.js/ai — ai.realtime() duplex voice sessions and ai.video() text-to-video. The heavy, demand-gated modalities, kept out of core so text/image/speech stay dependency-light.

BGOS channel adapter for Gobot: WS + REST client, outbound modalities, pair CLI, one-paste setup

DSH plugin: auto-fill metadata (context length, output cap, display name, input modalities) for custom-provider models from the Model Metadata Registry (models.dev).

DeepSeek Harness plugin: the Models+ settings page — a fully plugin-owned model/provider UI with per-model extension fields (reasoning efforts, input modalities, compat flags) and models.dev metadata prefill

DeepSeek Harness Web plugin: per-model input-modality selector on the Models settings page for third-party (pi-ai) providers

Pi coding agent extension — interactive table of your scoped models with pricing, context window, thinking levels, modalities, and coding benchmarks

DSH plugin: when a provider's model sync returns only ids, automatically fill display name, context window, max output tokens, reasoning efforts, and input modalities from the models.dev catalog (cached, TTL-refreshed), and surface a settings page with to

Gamma-entrainment protocol research agent — compare modalities/dosing, grade evidence, verify signed session bundles (ruv-neural)

dsh web 插件「模型定制工作台」: per-route/per-model model customization UI (input modalities, thinking levels, extra request params) over the llm-pi-ai / llm-deepseek settings sections

Gamma-entrainment protocol research agent — compare modalities/dosing, grade evidence, verify signed session bundles (ruv-neural)

DSH web plugin: per-model thinking effort (reasoningEfforts) and image support (input modalities) for llm-pi-ai custom providers, plus a plus-version custom provider creator.

A highly configurable toolkit for commanding robots in mixed modalities with liveliness.

Enhances pi's /model selector with context window, cost, input modalities, protocol, and reasoning info

State-of-the-art Machine Learning for the web. Run 🤗 Transformers directly in your browser, with no need for a server!