RESEARCH · RESEARCH · #3569
EvoX: Rust-based multi-model agent runtime with 'Second Policy' and Contrast (arXiv:2611.15006v1)
EvoX is a Rust-based, multi-model agent runtime proposing The Second Policy—treating the harness as an operational policy complementary to a model's semantic policy—and combining Action Realization, Execution Continuity, and Outcome Grounding. Evaluated with the Contrast framework across five benchmarks, six foundation models, and six harnesses, EvoX achieved a 56.2% success rate over 691 task–model pairs, outperforming the average of the other harnesses by ≈3.0 percentage points and showing larger gains on GLM-5.3 (+10.7) and Kimi K3 (+7.9); results were obtained without memory, evolution, or extensions.
KEY POINTS
- EvoX is a Rust-based, multi-model agent runtime proposing The Second Policy—treating the harness as an operational policy complementary to a model's semantic policy—and combining Action Realization, Execution Continuity, and Outcome Grounding.
- Evaluated with the Contrast framework across five benchmarks, six foundation models, and six harnesses, EvoX achieved a 56.2% success rate over 691 task–model pairs, outperforming the average of the other harnesses by ≈3.0 percentage points and showing larger gains on GLM-5.3 (+10.7) and Kimi K3 (+7.9); results were obtained without memory, evolution, or extensions.
- Highlights the importance of harness-level operational control and provides a reproducible runtime and evaluation approach that improves real-world tool-driven agent success across models and providers.
WHY IT MATTERS
Highlights the importance of harness-level operational control and provides a reproducible runtime and evaluation approach that improves real-world tool-driven agent success across models and providers.