b5-fusion
Multi-model B5 fusion ensemble for pi — every turn runs all candidates, the aggregator fuses the best answer
6.2k listings for pi
691 results · page 14 of 15
Multi-model B5 fusion ensemble for pi — every turn runs all candidates, the aggregator fuses the best answer
Multiple ChatGPT Codex accounts with aliases and weekly quota status for Pi
Automatically route work across models (planner/validator/executor/tool-parser) to produce the best possible output in pi. Quality-first orchestration with plan/agent/ask/debug modes.
LLM budget alerts: track spend, alert limits, prevent bill shock
Amp-style low/medium/high/ultra capability dial for Pi — pair an agent and oracle model per tier, mix providers, with multi-provider fallback chains
Pi image generation tool that auto-detects available API keys
Pi extension that adds prompt_cache_key to OpenAI Responses API requests for better prompt caching.
Pi provider plugin for Inception Labs (Mercury) — OpenAI-compatible chat completions with dynamic model discovery
A minimal pi extension that enables OpenAI Codex Fast mode via OPENAI_FAST_MODE=1.
Kilo provider extension for Pi. Access 300+ AI models through the Kilo Gateway.
Pi extension that adds a 2-line footer with live OpenAI Codex 5h/7d quota usage and reset timers.
Agentic engineering toolkit. Token estimation, cost, patterns, templates.
Pi extension adding Moonshot AI provider with Kimi K2.6, K2.5 and other models
Configurable Fast Mode toggle for OpenAI Responses and Codex models in pi.
Pi extension that adds xAI Grok OAuth provider and xAI utility tools.
Show AI cost in your local currencies — extends the pi footer with real‑time exchange rates from frankfurter.dev
Pi extension that shows Ollama Cloud quota usage in the status line
StreamLake AI custom provider extension for pi. Registers two provider entries (streamlake-coding / streamlake-payg) with baked-in base URLs, no redundant billing prompt, and dynamic model list fetch from the API.
Umans provider extension for pi coding agent. (No extra tools, no vision proxy, just the provider.)
Pi extension that supervises local llamafile-served model processes — start, stop, adopt, with progress visible on quit
A pi extension that tracks tool, skill, and extension usage stats
Pi extension that shows live Codex plan usage: remaining 5h and weekly limits, reset times, and cached fallback snapshots.
Pi extension: interactive multi-dimensional drill-down of local token usage (time / project / model)
Quickly switch models within the active pi-ccs provider.
Pi extension to view precise Google Cloud Code Assist (Gemini CLI) quotas.
Pi provider extension for Apfel's local Apple FoundationModels server.
A pi package that injects OpenAI's native web_search tool into the provider request payload.
Pi usages extension - Enhanced quota tracking with persistent overlay and multi-provider support
Create reusable Pi provider aliases for separate logins.
Lightweight Cursor model provider for Pi using one-shot Cursor Local Agent runs
Nexus Router provider for pi — private OpenAI-compatible inference via nexus-projects.ai
Animated input/output token odometers for pi coding agent.
CrofAI provider for Pi — self-contained, auto-discovers models, supports env var or /login paste
Pi extension that pins the current provider, model, and thinking level to project settings and reports local overrides on startup.
Pi extension that registers Inceptron (https://www.inceptron.io) as an LLM provider.
Configurable dynamic model router for Pi. Registers a virtual provider that routes each request to the best already-configured model for the situation.
pi extension for the Synthetic model provider
Pi extension that shows the Wayfinder router's model decision after each routed turn
Daily spend tracker for the Pi coding agent — shows total API cost across all of today's sessions in the footer status bar
Pi extension that loads Claude, Gemini, and Codex reusable command files as Pi commands and a model-callable tool.
A Pi footer extension for Codex subscription limit tracking.
Displays session and project token usage across Pi sessions.
Official Openference provider for Pi — device-flow OAuth login, live model discovery, and plan-aware pacing
MuleRouter provider for Pi — auto-discovers LLM models via the MuleRouter API
Tracks time usage, prompts, turns, and tool calls in a session.
pi extension for syncing providers/models from the Adesso AI Hub and viewing usage spend
Poolside AI provider extension for Pi - registers Poolside models as OpenAI-compatible provider
Long-answer HTML export extension for Oh My Pi and Pi.