ChorusGraph is a native agent graph runtime (Apache-2.0) — BSP scheduler plus a local semantic cache that skips redundant LLM calls when similar routing already happened. One pip install; not a LangGraph wrapper. this is an alternative with amazing benefits over LangGraph! https://insightitsgit.github.io/ChorusGraph/benchmarks.html see the numbers and never use langgraph to save your llm cost. https://insightitsgit.github.io/ChorusGraph/

Hey Product Hunt — I'm Amin, maker of ChorusGraph at Insight IT Solutions. Quick context: we're relaunching today as ChorusGraph — yesterday's listing pointed at the wrong product. This page is the native agent runtime, not our separate PrismGuard firewall. What ChorusGraph is: Apache 2.0 agent graph runtime — not a LangGraph wrapper. BSP engine + PrismCache semantic routing to skip redundant LLM calls, plus a native Route Ledger (per-hop audit without LangSmith). We publish paired benchmarks vs LangGraph — same prompts, tools, model on Azure: → Finance single (n=300 heavy): ~73% lower mean latency, ~76% fewer LLM calls → Healthcare multi: +15pp success, ~79% cache hits, abstain 3% vs 15% → We disclose where we tie: HC2 p95 wall-clock ~even at heavy scale Try it: pip install chorusgraph chorusgraph-demo Chroma RAG walkthrough: github.com/insightitsGit/ChorusGraph/tree/master/chorusgraph/examples/chroma_local_rag Full report: github.com/insightitsGit/ChorusGraph/blob/master/benchmark/results/azure_heavy_20260708_140300/mvp_scenarios/heavy_20260708_140300/COMPARISON_REPORT.md I'd love feedback on: 1. Where ChorusGraph fits vs your current LangGraph graphs 2. Cache hit patterns on your real agent traffic 3. What you'd need to pilot this on one production workflow I'll be here all day — ask anything.

How does the semantic cache actually detect similarity across different prompt phrasings, and what's the hit rate you're seeing in real workloads versus the benchmarks?

pip install was painless and the semantic cache actually cut repeat routing calls on my agent loop, pretty noticeable drop in token spend.

How does the semantic cache actually detect similarity across different prompt phrasings, and what's the hit rate you're seeing in real workloads versus the benchmarks?

pip install was painless and the semantic cache actually cut repeat routing calls on my agent loop, pretty noticeable drop in token spend.

ChorusGraph

Native agent runtime with semantic cache one pip install

Native agent runtime with semantic cache one pip install