PulseAugur
EN
LIVE 06:36:02

Kimchi AI agent outperforms DeepSeek Harness by 22 points in benchmark

By PulseAugur Editorial · [1 sources] ·

Kimchi, an AI coding agent, outperformed DeepSeek Harness by 22.3 percentage points in a benchmark comparison using the GLM 5.3 Flash model. The test, conducted by Kimchi's creator Matas Kaminskas, aimed to compare the harness frameworks themselves rather than the underlying models. Across three presets of DeepSeek Harness, the pass rates ranged from 44.0% to 47.1%, while Kimchi achieved a 69.4% pass rate. This finding suggests that the agent's framework significantly impacts performance, a pattern observed in previous comparisons. AI

IMPACT Highlights the significant impact of agent framework design on AI performance, suggesting that harness efficiency is a key factor in task completion rates.

RANK_REASON Comparison of AI agent harness performance using a benchmark. [lever_c_demoted from research: ic=1 ai=1.0]

Read on dev.to — LLM tag →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Kimchi AI agent outperforms DeepSeek Harness by 22 points in benchmark

How we ranked this

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
Comparison of AI agent harness performance using a benchmark. [lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
product, other
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
3 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

Full methodology in our editorial standards.

COVERAGE [1]

  1. dev.to — LLM tag TIER_1 English(EN) · Kimchi ·

    We ran DeepSeek Harness against Kimchi on the same model. It lost by 22 points

    <p><em>Originally published on the <a href="https://kimchi.dev/blog/deepseek-harness-vs-kimchi-terminal-bench" rel="noopener noreferrer">Kimchi blog</a> by Matas Kaminskas.</em></p> <p>DeepSeek Harness (dsh) came out on August 13, 2026. Seven weeks later it has 239,003 GitHub sta…

RELATED ENTITIES

RELATED TOPICS