Kimi
Kimi K3 (low)
Quality Score59.7
Value Score38.1
Reliability52.0
Cache Discount90%
Model specification
- Reasoning
- reasoning
- Input modalities
- text, image
- Output modalities
- text
- Context window
- 1,048,576 tokens
- Weights
- open weights
Token prices USD per 1M tokens
- Input
- $3.00
- Output
- $15.00
- Cached Input
- $0.300
- Cached Output
- not available
- Cost per task
- $0.242
Capability
- Intelligence
- 48.3
- Coding
- 72.0
- Agentic
- 39.6
- Omniscience
- 3.9
- Correct
- 45.8%
- Blended price
- $2.31
Answer outcomes
Correct 45.8%Incorrect 41.8%Abstained 12.4%
Artificial Analysis benchmarks
- GDPval-AA v2
- 38%
- τ³-Banking
- 42%
- Terminal-Bench v2.1
- 82%
- SciCode
- 51%
- Humanity’s Last Exam
- 25%
- GPQA Diamond
- 84%
- CritPt
- 3%
- AA-Omniscience
- 52%
- AA-LCR
- 77%
Similar models
- Kimi K2.7 CodeKimiQuality Score 49.2Value Score 38.4
- Kimi K3 (max)KimiQuality Score 69.6Value Score 45.7
- Kimi Linear 48B A3B InstructKimiQuality Score not availableValue Score not available
- Gemini 3.5 Flash (medium)GoogleQuality Score 60.3Value Score 41.6
- GPT-5.6 Terra (high)OpenAIQuality Score 58.6Value Score 40.1
- GPT-5.3 Codex (xhigh)OpenAIQuality Score 58.6Value Score 39.0