MoonshotAI: Kimi K3 (batch)
The Kimi K3 (batch) from MoonshotAI is a versatile AI model that can handle text, image, and video input modalities. It has a high context length of 1 MB, allowing it to process complex tasks and retain contextual information. For users who need a robust model with strong reasoning capabilities and support for various tools, the Kimi K3 (batch) is worth considering. Its price point of $3/M input tokens and $15/M output tokens may be steep for some, but its blended benchmark score of 96.2 across three independent benchmarks suggests it could deliver high-quality results in certain applications, making it a viable option for those with sufficient budget and specific needs.
Benchmark results
Independent, published benchmarks. Blended score 96.2 across 3 benchmarks, last refreshed 2026-09-04. 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:batch
- Vendor
- moonshotai
- Released
- July 2026
- Tokenizer
- Other
- Input Modalities
- text, image, video
- Output Modalities
- text
- Max Output
- 943,718 tokens
- 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 |
Category rankings
Where MoonshotAI: Kimi K3 (batch) places across the 42 categories it ranks in. How we rank →
| # | Category | Score |
|---|---|---|
| #6 | Video SummarizationVideo · of 25 ranked | 162 |
| #14 | Function / Tool CallingAgents · of 25 ranked | 156 |
| #15 | Browser AutomationAgents · of 25 ranked | 176 |
| #16 | Diagram ExtractionVision · of 25 ranked | 151 |
| #16 | Screenshot DebuggingVision · of 25 ranked | 145 |
| #16 | Chart & Graph ReadingVision · of 25 ranked | 155 |
| #17 | Data AnalysisData · of 25 ranked | 175 |
| #18 | Frontend Component DesignCode · of 25 ranked | 149 |
| #18 | Fitness CoachingPersonal · of 25 ranked | 137 |
| #20 | OCR / Document ParsingData · of 25 ranked | 147 |
| #20 | Table Extraction from PDFsData · of 25 ranked | 147 |
| #20 | Financial AnalysisProfessional · of 25 ranked | 163 |
| #20 | Math TutoringEducation · of 25 ranked | 150 |
| #20 | Experiment DesignResearch · of 25 ranked | 160 |
| #21 | Code ReviewCode · of 25 ranked | 170 |
| #21 | Unit Test GenerationCode · of 25 ranked | 157 |
| #21 | CI/CD PipelinesCode · of 25 ranked | 147 |
| #21 | Essay GradingEducation · of 25 ranked | 170 |
| #22 | Bug FixingCode · of 25 ranked | 185 |
| #22 | ETL ScriptingData · of 25 ranked | 157 |
| #22 | Medical Note SummarizationProfessional · of 25 ranked | 157 |
| #22 | Physics TutoringEducation · of 25 ranked | 150 |
| #22 | History TutoringEducation · of 25 ranked | 147 |
| #22 | Agent WorkflowsAgents · of 25 ranked | 197 |
| #22 | Coding AgentsAgents · of 25 ranked | 197 |
| #22 | Scientific CodingResearch · of 25 ranked | 185 |
| #23 | SQL GenerationCode · of 25 ranked | 173 |
| #23 | Code RefactoringCode · of 25 ranked | 167 |
| #23 | Creative WritingWriting · of 25 ranked | 160 |
| #23 | Screenwriting & DialogWriting · of 25 ranked | 160 |
| #23 | Academic WritingWriting · of 25 ranked | 160 |
| #23 | Legal DraftingProfessional · of 25 ranked | 160 |
| #23 | Legal ResearchProfessional · of 25 ranked | 170 |
| #23 | Contract ReviewProfessional · of 25 ranked | 180 |
| #23 | Scientific ResearchProfessional · of 25 ranked | 170 |
| #23 | RFP ResponseBusiness · of 25 ranked | 176 |
| #23 | OKRs & Strategic PlanningBusiness · of 25 ranked | 160 |
| #23 | RAG PipelinesAgents · of 25 ranked | 167 |
| #23 | Long-Context Q&AAgents · of 25 ranked | 157 |
| #23 | Fiction CollaboratorPersonal · of 25 ranked | 170 |
| #23 | Math ProofsResearch · of 25 ranked | 160 |
| #23 | Literature ReviewResearch · of 25 ranked | 170 |
Similar models
MoonshotAI: Kimi K3
MoonshotAI: Kimi K2.7 Code
MoonshotAI: Kimi K2.6
MoonshotAI: Kimi K2.5
MoonshotAI: Kimi K2 Thinking
MoonshotAI: Kimi K2 0905
Quick answers
- How much does MoonshotAI: Kimi K3 (batch) cost?
- $3.00 per million input tokens and $15.00 per million output tokens.
- What is MoonshotAI: Kimi K3 (batch)'s context window?
- 1,048,576 tokens, roughly 1,572 pages of text in a single request.
- Does MoonshotAI: Kimi K3 (batch) support tool calling?
- It supports tool calling, structured output, a reasoning mode.
- Can MoonshotAI: Kimi K3 (batch) process images?
- Yes, it accepts image input.