From 9dd56f0dfb040342b9fb1ec78ff255c5854b4063 Mon Sep 17 00:00:00 2001
From: Teknium <127238744+teknium1@users.noreply.github.com>
Date: Fri, 26 Jun 2026 11:05:07 -0700
Subject: [PATCH] docs(moa): add HermesBench results to Mixture of Agents page
 (#53206)

---
 .../docs/user-guide/features/mixture-of-agents.md    | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/website/docs/user-guide/features/mixture-of-agents.md b/website/docs/user-guide/features/mixture-of-agents.md
index 9dbe3e65583..c901a2fe9e7 100644
--- a/website/docs/user-guide/features/mixture-of-agents.md
+++ b/website/docs/user-guide/features/mixture-of-agents.md
@@ -106,6 +106,18 @@ hermes moa configure review       # create or update a named preset
 hermes moa delete review
 ```
 
+## Benchmarks
+
+On HermesBench, a two-model MoA preset — `claude-opus-4.8` aggregating over a `gpt-5.5` reference — outscores either model run on its own:
+
+| Model | HermesBench score |
+|---|---|
+| **Opus aggregator (opus-4.8 + gpt-5.5 reference) — MoA** | **0.8202** |
+| `anthropic/claude-opus-4.8` | 0.7607 |
+| `openai/gpt-5.5` | 0.7412 |
+
+The MoA configuration beats its strongest component (opus-4.8) by ~6 points, confirming that aggregating a second perspective lifts quality on hard tasks rather than just averaging the two.
+
 ## Notes
 
 - MoA is no longer listed under `hermes tools`; there is no `moa` toolset to enable.