Choose an institution to see its records and dossier, or choose a topic.
Through year 2027 · all records, including scheduled dates 2026 2025 2024 2023 2022 2021 2020 2019 2018 2017 2016 2015 2014 2013 2012 2011 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999 1998 1997 1996 1994 1993 1992 1991 1989 1986 1985 1984 1983 1979 1966 1965 1961 1960 1955 1952 1950 1872 1863 Show Organisations and infrastructure All types Organisation Person Document Concept Infrastructure Public body Reset filters
100 people, organisations and other records · 129 relationships · 34 claims about a valueEverything recorded up to the chosen year, whether or not still in effect · Undated records included
Claim and evidence Close ×
The Kimi K2 technical report states that the team conducted red-teaming evaluations of Kimi K2 against other open-source models, using Promptfoo to generate adversarial prompts covering harmful, privacy and security content. Copy claim link
Copy claim reference
Machine-checked, not reviewed by an editor Stated in source Supported
Supporting evidence: primary · 1 source
Date 2025-07-28 · publication Scope and limits What the report states about its own evaluation; the results and their adequacy are not recorded. The report states that, to minimise subjectivity, it conducted multiple rounds of human review and assigned the same reviewer to all cases in a given test set. Not evidence of compliance with any commitment.
The evidence 02 primary source Supports the claim
Kimi K2: Open Agentic Intelligence (technical report, v1) ↗ arXiv · 2025-07-28
Copy source reference
Cite the original source Cite the original document, not Selekoa. Check these details against it.
Copy APA
Download BibTeX Download RIS Download CSL-JSON
Source excerpt “We conducted red-teaming evaluations on Kimi K2 compare with other open-source LLMs.” Location in source Section 4.3.1, “Experiment Settings”
How it was read Retrieved directly from the publisher
Machine-checked · 2026-10-27
Source details and check notes Authors Kimi Team Version Dated document, retrieved on 2026-10-27 (SHA-256 b838917cf0a475d2e1b2f1ba4d0b4c91b6f6c8be95f15aa0d4f3ea76e4ef60f5) directly from the publisher. Version: 2507.20534v1. PDF of version 1; later versions not inspected. Canonical address Open the canonical source ↗ Accessed 2026-10-27 Language English (en) Attribution Kimi Team, “Kimi K2: Open Agentic Intelligence (technical report, v1)”, 2025-07-28. Reuse No open redistribution licence asserted. Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution. Check notes Candidate pack passage, Chinese frontier-AI institutions, second collection (research/china-labs-2). The quotation (12 words) matched the retrieved text on 2026-10-27 and again in a fresh retrieval the same day. Read directly from the publisher. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-27, item K044) re-read the record’s sources and judged it partly supported; its corrections were applied at admission. No human editorial review. primary source Supports the claim
Kimi K2: Open Agentic Intelligence (technical report, v1) ↗ arXiv · 2025-07-28
Copy source reference
Cite the original source Cite the original document, not Selekoa. Check these details against it.
Copy APA
Download BibTeX Download RIS Download CSL-JSON
Source excerpt “We choose Promptfoo5 to generate adversarial prompts and analyze the responses.” Location in source Section 4.3.1
How it was read Retrieved directly from the publisher
Machine-checked · 2026-10-27
Source details and check notes Authors Kimi Team Version Dated document, retrieved on 2026-10-27 (SHA-256 b838917cf0a475d2e1b2f1ba4d0b4c91b6f6c8be95f15aa0d4f3ea76e4ef60f5) directly from the publisher. Version: 2507.20534v1. PDF of version 1; later versions not inspected. Canonical address Open the canonical source ↗ Accessed 2026-10-27 Language English (en) Attribution Kimi Team, “Kimi K2: Open Agentic Intelligence (technical report, v1)”, 2025-07-28. Reuse No open redistribution licence asserted. Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution. Check notes Candidate pack passage, Chinese frontier-AI institutions, second collection (research/china-labs-2). The quotation (11 words) matched the retrieved text on 2026-10-27 and again in a fresh retrieval the same day. Read directly from the publisher. Matching ignored differences of whitespace, case, quotation marks, apostrophes, dashes and Unicode compatibility forms, so the source’s typography can differ. A cross-family machine check (2026-10-27, item K044) re-read the record’s sources and judged it partly supported; its corrections were applied at admission. No human editorial review. Status and meaning Editorial status. Not reviewed by an editor . An exploratory record that no editor has accepted. This status does not say whether a person has read any linked passage. Each evidence record above shows how it was checked.
Relation. Publication of a specific document or version. Publication of a policy is not evidence of implementation.
Date. Statement in arXiv version 1; later versions not inspected.
Record ID c2-moonshot-k2-report-redteam