Kimi K2.5 is Moonshot AI's native multimodal model, delivering state-of-the-art visual coding capability and a self-directed agent swarm paradigm. Built on Kimi K2 with continued pretraining over approximately 15T mixed...
Tested on 27 benchmarks with 52.0% average. Top scores: OpenCompass — IFEval (93.9%), OTIS Mock AIME 2024-2025 (92.2%), OpenCompass — AIME2025 (91.9%).
Qwen3 235B A22B Thinking 2507 scores 59.4 (100% as good) at $0.15/1M input · 66% cheaper
OpenCompass Live Code Bench v6. Fresh competitive programming problems to evaluate code generation without memorization.
Real-world software engineering tasks from GitHub issues. Models must diagnose bugs and write patches that pass test suites. Human-verified subset of SWE-bench.
Unusual and adversarial machine learning challenges. Tests robustness of reasoning about edge cases in ML systems.
Abstraction and Reasoning Corpus. Tests fluid intelligence through novel visual pattern recognition puzzles. Core measure of general intelligence.
Deceptively simple questions that humans find easy but AI models often get wrong. Tests common sense and reasoning gaps.
ARC-AGI 2, harder sequel to ARC. More complex abstract reasoning patterns that test generalization ability beyond training data.
Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.
OpenCompass evaluation on AIME 2025 problems. Tests mathematical reasoning on fresh competition problems.
Original research-level math problems created by professional mathematicians. Problems are unpublished and cannot be memorized.
- Typemultimodal
- Context262K tokens (~131 books)
- ReleasedJan 2026
- LicenseOpen Source
- StatusActive
- Cost / Message~$0.003