Expired75% confidenceFactTime unknown
In Velse AI's latest rankings on April 23, Kimi K2.6 ranked first and DeepSeek V4 ranked second, with only a 0.07 percentage point difference between the two.
1
Sources
75%
Confidence
Medium-term (~90 days)
Relevance
5/31/2026
First Seen
Valid until: 8/29/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedThe performance gap between Kimi K2.6 and Claude on SWE-Bench is less than 1%, but the price difference is 7x (Kimi is $0.8 per million tokens, Claude is $5).65% similarUnverifiedIn the overall performance rankings, Kimi K2.5 scored 64%, placing fifth, with the top four being Gemini 3 Pro (100%), Claude Opus 4.5 Max (74%+), GLM 4.7 (65%), and GPT-5.2X (65%).65% similarUnverifiedKimi K2.5's token usage is 50% higher than that of the second- and third-ranked Gemini and Claude small models.64% similarUnverifiedK2.7 Code has not yet published its SWE-Bench scores, and all performance data is self-reported by Moonshot AI; the previously released K2.6 had a self-reported score that differed from third-party reproductions by 15 to 20 percentage points.63% similarUnverifiedIn an independent tester's overall rankings, Claude 4.5 Haiku placed 34th and 37th respectively, making it the lowest-scoring next-generation model among mainstream AI labs.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/13852API
curl https://kongchang.com/api/v1/knowledge/claims/13852MCP
get_claim(id=13852)