Home/Events/CUGA vs Hermes and Deep Agents: Performance Comparison on AppWorld Tasks

CUGA vs Hermes and Deep Agents: Performance Comparison on AppWorld Tasks

Confirmed

Confidence

90%

Impact: 80%

Updated 3h ago

Consensus Brief

The article compares the performance of three AI agent harnesses—CUGA, Hermes, and Deep Agents—using the same GLM-5.3-flash model across 90 tasks in AppWorld. CUGA demonstrated superior task completion rates, achieving 81.1% compared to Hermes at 46.7% and Deep Agents at 30.0%. The findings emphasize the importance of selecting the appropriate harness for specific workloads in enterprise environments.

Sourced from

Hugging FaceT1 Hugging FaceT1 Hugging FaceT1+2 more

Primary: Hugging Face

What Changed Since Last Update

3h ago

New official source added: Hugging Face published an update on Tue, 03 No ("Same Model, Different Harness: CUGA, Hermes, and Deep Agents — When to Pick CUGA?").

Claim Ledger

4 claims tracked across sources

Confirmed Fact

CUGA achieved 81.1% task completion on AppWorld tasks.

Confirmed Fact

Hermes achieved 46.7% task completion on AppWorld tasks.

Confirmed Fact

Deep Agents achieved 30.0% task completion on AppWorld tasks.

Official Claim

CUGA is designed for enterprise use with specific rule enforcement capabilities.

Role-Based Impact Analysis

Source Timeline

3 sources corroborating

Same Model, Different Harness: CUGA, Hermes, and Deep Agents — When to Pick CUGA?

Hugging Face·3h ago

Same Model, Different Harness: CUGA, Hermes, and Deep Agents — When to Pick CUGA?

Hugging Face·3h ago

Same Model, Different Harness: CUGA, Hermes, and Deep Agents — When to Pick CUGA?

Hugging Face·3h ago