Expired60% confidenceOpinionTime unknown
In front-end development tests, Kimi K2.6 surpassed Claude Opus 4 in certain scenarios.
2
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
5/31/2026
First Seen
Valid until: 8/29/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedKimi K2.6 achieved a score of 58.6% on the SWE-Bench Pro benchmark, ranking first among open-source models.72% similarUnverifiedKimi K2.6 competes with closed-source models such as GPT 5.4, Claude Opus 4.6, and Gemini 3.1 Pro on top benchmarks including SWE Bench Pro, deep search question answering, and Humanity's Last Exam.72% similarUnverified以Kimi K3实验，给它预填充1%的Claude Opus解码推理片段，其输出风格就会明显向Opus靠拢，某些区段Claude推理链从Kimi K3提取的容易程度比其他模型高出约6个数量级69% similarVerifiedIn the Terminal Bench 2.0 test, K2.6 scored 66.7, slightly higher than both GPT-5.4 and Claude Opus 4.6.68% similarUnverifiedClaude Opus 4.6 leads in performance on real-world expert-level tasks in the SWE-bench benchmark.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/3980API
curl https://kongchang.com/api/v1/knowledge/claims/3980MCP
get_claim(id=3980)