nowfound

AI · November 18, 2026

Werkbank – 40 tool-using tasks to evaluate self-hosted LLMs

Autonomy benchmark for self-hosted LLM agents. Verifiable task completion, stateful workflows, coding, and efficiency. - aisyndicate-ch/victors-werkbank

Werkbank – 40 tool-using tasks to evaluate self-hosted LLMs launched on November 18, 2026 with 3 votes, #861 of 3,085 launches that month and more than 50% of that year's launches.

What it does

Autonomy benchmark for self-hosted LLM agents. Verifiable task completion, stateful workflows, coding, andfrom github.com

Does a similar job

all alternatives →
  • Microsoft-Decision-1Oct 2026 · commandline.microsoft.com · ▲174

    Microsoft’s decision model for agents and workflows

  • 20+ Claude Code agents coordinating on real work (open source)Feb 2026 · github.com · ▲53

    Single-agent LLMs suck at long-running complex tasks. We’ve open-sourced a multi-agent orchestrator that we’ve been using to handle long-running LLM tasks. We found that single LLM agents tend to stall, loop, or generate non-compiling code, so we built a harness for agents to coordinate over shared context while work is in progress. How it works: 1. Orchestrator agent that manages task decomposition 2. Sub-agents for parallel work 3. Subscriptions to task state and progress 4. Real-time sharing of intermediate discoveries between agents We tested this on a Putnam-level math problem, but the…

  • Backenly20d ago · backenly.com · ▲169

    The autonomous backend for AI coding agents

  • APIEval-20May 2026 · resources.kusho.ai · ▲121

    An open benchmark for AI agents that test APIs

  • Praex30d ago · praex.ai · ▲68

    The coding agent that runs on its own AI models

  • ZUS9d ago · zus.8syncdev.com · ▲75

    AI-native IDE: agents code with human review

More ai this month

the category →

Launched alongside, November 2026

the whole month →