Moonshot AI
kimi-k2.5
mitMultimodal
Benchmark Average (0–100)
77.3%
Average of non-null benchmark scores across all evaluated tasks.
BENCHMARK SCORES
GPQA87.6%
Graduate-Level Google-Proof Q&A. PhD-level multiple-choice questions in chemistry, biology, and physics. Scored by accuracy.
AIME 202596.1%
AIME 2025 is a mathematics competition benchmark testing advanced problem-solving.
SWE-bench Verified76.8%
SWE-bench Verified evaluates models on real-world software engineering tasks from GitHub issues.
BrowseComp74.9%
BrowseComp tests web browsing comprehension, measuring ability to find and synthesize information from websites.
HLE50.2%
Humanity's Last Exam (HLE) is a multi-modal benchmark testing frontier knowledge across mathematics, humanities, and natural sciences with 2,500 expert-level questions.
MMMU-Pro78.5%
MMMU-Pro is a more challenging version of MMMU with harder multimodal understanding tasks.
OTHER BENCHMARKS
CharXIV-R
77.5%
MODEL INFO
Organization
Moonshot AI
Context Window
262,100
Release Date
2026-01-27
Knowledge Cutoff
—
Parameters
1000000000000.0B
License
mit
Input Price ($/M)
$0.6000
Output Price ($/M)
$2.5000
METADATA
Announcement Date
2026-01-27
Organization Country
CN
Canonical ID
—