# Kodezi/Chronos

Kodezi Chronos is a debugging-first language model that achieves state-of-the-art results on SWE-bench Lite (80.33%) and 67% real-world fix accuracy, over six times better than GPT-4. Built with Adaptive Graph-Guided Retrieval and Persistent Debug Memory. Model available Q1 2026 via Kodezi OS.

Repository: https://github.com/Kodezi/Chronos
Canonical: https://ross.abutalabs.com/products/chronos
Homepage: https://chronos.so/
Language: Java
License: NOASSERTION
License Family: other
Topics: artificial-intelligence, benchmark, benchmark-report, bug-fixing, chronos, code, code-analysis, code-analysis-tool, code-debugger, code-understanding, debugging, developer-tools, kodezi, language-model, machine-learning, program-repair, software-engineering, autonomous-debugging
Last push: 2025-11-12T09:53:44+00:00

## Health v2 (maintenance only)
Score: 41/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 51, release rhythm 35, longevity 29
- inputs: {"age_days": 408, "days_push": 294, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4922, forks 210 (observed 2026-08-28T04:09:02.476419+00:00)

## What it is
Kodezi Chronos is a proprietary debugging-first language model for repository-scale code understanding and autonomous bug fixing, presented here via a research paper, benchmark results, and leaderboards. The repository itself contains no runnable code - the model is only accessible through Kodezi OS (beta Q4 2025, GA Q1 2026).

## Use cases
- automatically fix bugs in large codebases
- evaluate LLM performance on SWE-bench Lite
- research debugging-focused language models
- compare program repair model results
- find state-of-the-art autonomous debugging benchmarks
- understand graph-guided retrieval for code models

## When to choose
- you want to read benchmark results or the research paper on debugging LLMs
- you are evaluating SWE-bench Lite leaderboard claims
- you plan to access the model via Kodezi OS when it ships

## When to avoid
- you need runnable open-source code or model weights today
- you want a self-hostable debugging tool
- you need an MIT/Apache-licensed dependency - the license is non-standard and the model is proprietary

## Facets
- artifact type: learning-resource
- maturity: experimental
- function: machine-learning, llm-training, developer-tools
- domain: artificial-intelligence, large-language-models, developer-tools, programming-languages
- platform: python, cloud
- tags: debugging-llm, program-repair, swe-bench, benchmark-results, research-paper, proprietary-model, code-understanding, autonomous-debugging

## Member repositories
- Kodezi/Chronos (main) score 41

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:02.476419+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:18:01.336024+00:00, confidence not recorded.
  - readme: https://github.com/Kodezi/Chronos (fetched 2026-08-28T04:09:02.476419+00:00, sha f00a1ad45403)
- Data as of 2026-08-30T08:39:29.467469+00:00.