MoonshotAI: Kimi K3
MoonshotAI's Kimi K3 is a versatile AI model capable of processing text, images, and videos within an extensive context length of 1048576 tokens. It supports both reasoning and the use of tools, making it suitable for complex tasks that require integration with external data or actions. Given its benchmark blended score of 96.9 across three independent evaluations, Kimi K3 is a strong contender for projects demanding high performance in multiple domains. However, at a price point of $3.0 per million input tokens and $15.0 per million output tokens, it may be cost-prohibitive for smaller-scale applications or those with limited budgets. Organizations requiring robust AI capabilities across various modalities should consider whether the investment aligns with their needs.
Benchmark results
Independent, published benchmarks. Blended score 96.9 across 3 benchmarks, last refreshed 2026-08-13. How scoring works →
| Benchmark | Measures | Score |
|---|---|---|
| AI Index | broad capability composite | 98.5 |
| AI Index Coding | software engineering tasks | 100.0 |
| AI Index Agentic | multi-step tool-using tasks | 89.5 |
- Model ID
- moonshotai/kimi-k3
- Vendor
- moonshotai
- Released
- July 2026
- Tokenizer
- Other
- Input Modalities
- text, image, video
- Output Modalities
- text
- Max Output
- default
- Tool Calling
- ✓ supported
- Structured Output
- ✓ supported
- Reasoning Mode
- ✓ supported
- Vision
- ✓ accepts images
- Audio
- no
- Moderated
- no
What it costs in practice
Computed from the current $3.00/M input and $15.00/M output rates. Run your own numbers →
| Job | Tokens | Cost |
|---|---|---|
| Summarize a 50-page report | 30k in / 1.5k out | $0.11 |
| Classify 1,000 customer emails | 500k in / 50k out | $2.25 |
| A month of a busy support chatbot | 5M in / 2M out | $45.00 |
Strong choice for
Category rankings
Where MoonshotAI: Kimi K3 places across the 49 categories it ranks in. How we rank →
| # | Category | Score |
|---|---|---|
| #4 | Video SummarizationVideo · of 25 ranked | 162 |
| #9 | Function / Tool CallingAgents · of 25 ranked | 156 |
| #10 | Browser AutomationAgents · of 25 ranked | 176 |
| #11 | Diagram ExtractionVision · of 25 ranked | 151 |
| #12 | Data AnalysisData · of 25 ranked | 175 |
| #13 | Frontend Component DesignCode · of 25 ranked | 149 |
| #13 | Screenshot DebuggingVision · of 25 ranked | 145 |
| #13 | Chart & Graph ReadingVision · of 25 ranked | 155 |
| #13 | Fitness CoachingPersonal · of 25 ranked | 137 |
| #14 | Financial AnalysisProfessional · of 25 ranked | 163 |
| #15 | OCR / Document ParsingData · of 25 ranked | 147 |
| #15 | Table Extraction from PDFsData · of 25 ranked | 147 |
| #15 | Experiment DesignResearch · of 25 ranked | 160 |
| #16 | Math TutoringEducation · of 25 ranked | 150 |
| #17 | SQL GenerationCode · of 25 ranked | 173 |
| #17 | Unit Test GenerationCode · of 25 ranked | 157 |
| #17 | CI/CD PipelinesCode · of 25 ranked | 147 |
| #18 | Bug FixingCode · of 25 ranked | 185 |
| #18 | Physics TutoringEducation · of 25 ranked | 150 |
| #18 | Agent WorkflowsAgents · of 25 ranked | 197 |
| #18 | Coding AgentsAgents · of 25 ranked | 197 |
| #18 | Scientific CodingResearch · of 25 ranked | 185 |
| #19 | Code ReviewCode · of 25 ranked | 170 |
| #19 | Marketing CopyWriting · of 25 ranked | 131 |
| #19 | Essay GradingEducation · of 25 ranked | 170 |
| #19 | Standardized Test PrepEducation · of 25 ranked | 131 |
| #20 | ETL ScriptingData · of 25 ranked | 157 |
| #20 | Medical Note SummarizationProfessional · of 25 ranked | 157 |
| #20 | History TutoringEducation · of 25 ranked | 147 |
| #21 | Code RefactoringCode · of 25 ranked | 167 |
| #21 | Creative WritingWriting · of 25 ranked | 160 |
| #21 | Screenwriting & DialogWriting · of 25 ranked | 160 |
| #21 | Academic WritingWriting · of 25 ranked | 160 |
| #21 | Legal DraftingProfessional · of 25 ranked | 160 |
| #21 | Legal ResearchProfessional · of 25 ranked | 170 |
| #21 | Contract ReviewProfessional · of 25 ranked | 180 |
| #21 | Scientific ResearchProfessional · of 25 ranked | 170 |
| #21 | RFP ResponseBusiness · of 25 ranked | 176 |
| #21 | OKRs & Strategic PlanningBusiness · of 25 ranked | 160 |
| #21 | RAG PipelinesAgents · of 25 ranked | 167 |
| #21 | Long-Context Q&AAgents · of 25 ranked | 157 |
| #21 | Fiction CollaboratorPersonal · of 25 ranked | 170 |
| #21 | Math ProofsResearch · of 25 ranked | 160 |
| #21 | Literature ReviewResearch · of 25 ranked | 170 |
| #22 | Long-Document SummarizationWriting · of 25 ranked | 161 |
| #22 | Blog Post WritingWriting · of 25 ranked | 141 |
| #22 | Character RoleplayPersonal · of 25 ranked | 151 |
| #22 | Trip PlanningPersonal · of 25 ranked | 141 |
| #24 | SOP / Process DocsBusiness · of 25 ranked | 137 |
Similar models
MoonshotAI: Kimi K2.7 Code
MoonshotAI: Kimi K2.7 Code (batch)
MoonshotAI: Kimi K2.6
MoonshotAI: Kimi K2.5
MoonshotAI: Kimi K2 Thinking
MoonshotAI: Kimi K2 0905
Quick answers
- How much does MoonshotAI: Kimi K3 cost?
- $3.00 per million input tokens and $15.00 per million output tokens.
- What is MoonshotAI: Kimi K3's context window?
- 1,048,576 tokens, roughly 1,572 pages of text in a single request.
- Does MoonshotAI: Kimi K3 support tool calling?
- It supports tool calling, structured output, a reasoning mode.
- Can MoonshotAI: Kimi K3 process images?
- Yes, it accepts image input.