Kimi K2 Thinking
moonshot/kimi-k2-thinking- Moonshot AI
- Family · Kimi K2
- Released 6 Nov 2025
- Open weights
- Text → Text
- Tools · JSON
Not on OSMEXO yet
Spec & list price
Context250K
Max output96K
Input / 1M$0.60
Output / 1M$2.504.2× input
Scores · standing
Intelligence22.0#122/489
Usage
Tokens / day147.2M▲ 20% 7d
Usage rank#284
Requests / day13.0K
Tool calls / req0.30
Per request · 30 days
Avg input14.5Kabout median
Avg output3.0K4.4× median
Cache hit74%about median
Reasoning share87%1.5× median
Tool-call failures0.8%0.4× median
Consumption
Peak 1.8B 28 OctLow 67.6M 31 OctRank today #284
News & trends
6 mentions this week ▼ 60%Kimi K2 family · videos, posts, articles · 30 days
Trending · 14 days · Kimi K2 family
- Anthropic Haiku 5.5 vs Moonshot Kimi K2.6 draw “a beach scene” #shorts199 views
- Kimi k2.57 upvotes
- Kimi K2利用者は12月末までに移行可否を確認 #shorts281 views
- Stuck between GLM 4.7, 4.6, and Kimi K2.5. Also, which provider should I use between OpenRouter and DeepInfra?49 upvotes
News · latest from Moonshot AI
- ReplaySSM: Rethinking State Storage for Faster Hybrid-Model DecodingNVIDIA · 12 Oct
- Together Link: open models in the harness you already use. Start with one command today.Together AI · 5 Oct
- Kimi K3: the complete developer guideTogether AI · 1 Aug
- Together AI announces strategic partnership with Moonshot AI to natively serve Kimi modelsTogether AI · 29 Jul
Benchmarks
bar = share of the top score · Benchmark data: Artificial AnalysisSpeed · median across providers, latest day
Output speed#274/33439 tok/s1.7× slower than median
First token#2481.78 s1.7× longer than median
#21 cheapest of 31 with a similar score
Value
Intelligence Index
Blended price per 1M tokens (log)
Family progress
Intelligence Index
- Family releases27.0
Technical details
- Tokenizer
- Other
- License
- other
- Parameters
- 1T
- Moderation
- none
- Downloads · likes
- 47.6K · 1.7K
Request parameters 15
- tools
- tool_choice
- response_format
- structured_outputs
- reasoning
- include_reasoning
- temperature
- top_p
- top_k
- max_tokens
+5 more
- stop
- seed
- frequency_penalty
- presence_penalty
- repetition_penalty
When to use
Best for
- Math and competition-style problem solving
- Multi-step agent workflows with tool calls
- Hard coding and debugging
- Long-context reasoning over big documents
Think twice for
- Low-latency chat (thinks before answering)
- Image understanding (text-only input)
- High-volume simple tasks
Research papers
- Ideological Stance Detection in a Low-Resource Language: Polarization in Bangladeshi Public vs Private University Discourse on Social Media3 Oct
- Model Forensics: Investigating Whether Concerning Behavior Reflects Misalignment24 Jun
- Information Extraction of Nested Complex Structure of Quantum Cascade Lasers via Large Language Models11 May
- Beyond Task Completion: Revealing Corrupt Success in LLM Agents through Procedure-Aware Evaluation3 Mar
- KLong: Training LLM Agent for Extremely Long-horizon Tasks19 Feb
Same family
Kimi K2- 27.0
- 23.5
- 12.7
- 15.3
Alternatives
other vendors · similar capability- 26.3
- 21.8
- T25.0
- 21.5
Want Kimi K2 Thinking on OSMEXO?
Sign in to request it. Requests are reviewed; a request does not guarantee availability.