ModelsIndexCompareBest forBenchmarksStatusPricingAPI

Kimi K2.5

Moonshot AI·Operational·Released 2026-01-27textvisionvideo
Context256k
Max output-
Input / 1M-
Output / 1M-
Cached in-
Benchmarks26

Benchmarks

Cited public sources · rank is exact-benchmark, apples-to-apples

SAGE (Vals AI)#18/7749.9% ↗
SimpleQA Verified (Epoch AI run)#48/7134.3% ↗
SWE-bench Verified (Epoch AI run)#16/3173.8% ↗
Vending-Bench 2 (Andon Labs)#44/591198.46 ↗
WeirdML (Håvard Tveit Ihle)#56/9245.6% ↗
LMArena Elo#56/1411450.1 ↗
LMArena Vision (LMArena)#37/771252.2 ↗
LMArena WebDev (LMArena)#53/941435.8 ↗
Toolathlon-Verified (HKUST)#23/2333% ↗
SimpleBench (SimpleBench)#52/7346.8% ↗
OSWorld-Verified (XLANG)#8/1363.3% ↗
AIME 2025#3/2496.1% ↗
BrowseComp#14/2360.6% ↗
GPQA-Diamond#1/887.6% ↗
HLE-Full30.1% ↗
HLE-Full (w/ tools)50.2% ↗
HMMT 2025 (Feb)95.4% ↗
IMO-AnswerBench#1/281.8% ↗
LiveCodeBench (v6)#1/385% ↗
MMLU-Pro#4/4987.1% ↗
MMMU-Pro#6/1278.5% ↗
SWE-Bench Multilingual#1/273% ↗
SWE-Bench Pro#19/2150.7% ↗
SWE-Bench Verified#2/376.8% ↗
Terminal Bench 2.0#1/250.8% ↗
VideoMMMU#1/586.6% ↗

Compare

vs Gemini 4 Argonvs GPT-6.1 Solvs Claude Sonnet 5.5vs Muse Spark 1.3vs GPT-6 Solvs GPT-6 Lunavs Claude Opus 5.5vs GPT-6 Astra

More from Moonshot AI

Kimi K3Kimi K2.7 CodeKimi K2.6Kimi K2 ThinkingKimi K2-0905Kimi K2

Kimi K2.5 FAQ

What is Kimi K2.5?
Kimi K2.5 is a large language model from Moonshot AI, released on January 27, 2026. It accepts text, image and video input.
When was Kimi K2.5 released?
Moonshot AI released Kimi K2.5 on January 27, 2026.
What is Kimi K2.5's context window?
Kimi K2.5 has a 256K-token context window.
How does Kimi K2.5 score on benchmarks?
Kimi K2.5 has 26 cited benchmark scores, including SAGE (Vals AI) 49.9%, SimpleQA Verified (Epoch AI run) 34.3% and SWE-bench Verified (Epoch AI run) 73.8%. Each score links to its source.

Access Kimi K2.5 and every other model through one endpoint with automatic failover: Respan gateway.