Kimi K3
Kimi K3 is Moonshot AI (Kimi)'s language model with a 1.0M context window and up to 944K output tokens, available from 23 providers, starting at $0.192 / 1M input and $12.50 / 1M output. A flagship reasoning-capable multimodal LLM with 2.8 trillion parameters, built on Kimi Delta Attention (a hybrid linear attention mechanism) with native visual understanding and tool-use support.
| Specifications | |
|---|---|
moonshot-kimi-k3 | |
| Language | |
| Active | |
| Moonshot AI (Kimi) | |
| 1.0M tokens | |
| 944K tokens | |
| ImagePDFTextVideo | |
| Text | |
| default | |
| 2.78T | |
| 7,099 | |
| 2,850 | |
| 2,855 | |
| · 4 months ago | |
| Benchmarks | |
|---|---|
| Intelligence Index | #18 |
| Coding Index | #10 |
| GPQA | #11 |
| HLE | #22 |
| Time to First Token | #542 |
| SciCode | #12 |
| LCR | #1 |
| Output TPS | #185 |
Capabilities
Input4/5
Text✓
Image✓
Audio·
Video✓
PDF✓
Output1/5
Text✓
Image·
Audio·
Video·
Embedding·
Capabilities7/13
Reasoning✓
Adaptive Reasoning·
Function Calling✓
Parallel Function Calling✓
Structured Outputs✓
Native JSON Schema✓
Web Search✓
URL Context·
Computer Use·
Code Execution·
File Search·
Prompt Caching✓
Assistant Prefill·
Pricing by Provider
US Dollar ($)
Per 1M tokens
| Provider | Standard | Flex | Fast | Batch | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Input $ / 1M | Output $ / 1M | Cache Read $ / 1M | Cache Write 5m $ / 1M | Cache Write 1h $ / 1M | Input $ / 1M | Output $ / 1M | Cache Read $ / 1M | Cache Write 5m $ / 1M | Input $ / 1M | Output $ / 1M | Cache Read $ / 1M | Cache Write 5m $ / 1M | Input $ / 1M | Output $ / 1M | Cache Read $ / 1M | |
| $3.00 | $12.50 | $0.5 | N/A | N/A | — | — | — | — | — | — | — | — | — | — | — | |
| $3.00 | $15.00 | $0.3 | N/A | N/A | — | — | — | — | — | — | — | — | — | — | — | |
| $2.83 | $14.13 | N/A | N/A | N/A | — | — | — | — | — | — | — | — | $1.41 | $7.07 | N/A | |
| $3.30 | $16.50 | $0.3 | $3.75 | N/A | $1.65 | $8.25 | $0.15 | $1.88 | $5.78 | $28.88 | $0.525 | $6.56 | $1.65 | $8.25 | N/A | |
| $3.00 | $15.00 | $0.3 | N/A | N/A | — | — | — | — | — | — | — | — | — | — | — | |
| $3.00 | $15.00 | $0.3 | N/A | N/A | — | — | — | — | — | — | — | — | — | — | — | |
| $3.00 | $15.00 | $0.3 | $3.00 | N/A | — | — | — | — | $5.25 | $26.25 | $0.525 | $5.25 | — | — | — | |
| $2.85 | $14.25 | $0.285 | N/A | N/A | — | — | — | — | — | — | — | — | — | — | — | |
| $3.00 | $15.00 | $0.3 | N/A | N/A | — | — | — | — | $3.75 | $18.75 | $0.375 | N/A | — | — | — | |
| $3.00 | $15.00 | $0.3 | N/A | N/A | — | — | — | — | — | — | — | — | — | — | — | |
| $3.00 | $15.00 | N/A | N/A | N/A | — | — | — | — | — | — | — | — | — | — | — | |
| $3.05 | $15.26 | $0.405 | N/A | N/A | — | — | — | — | — | — | — | — | — | — | — | |
| $3.00 | $15.00 | $0.3 | $3.00 | $6.00 | — | — | — | — | — | — | — | — | — | — | — | |
| $3.00 | $15.00 | $0.3 | N/A | N/A | — | — | — | — | — | — | — | — | — | — | — | |
| $3.00 | $15.00 | $0.3 | $3.00 | N/A | — | — | — | — | — | — | — | — | — | — | — | |
| $0.192 | $15.00 | $0.182 | N/A | N/A | — | — | — | — | — | — | — | — | N/A | N/A | N/A | |
| — | — | — | — | — | — | — | — | — | — | — | — | — | $2.28 | $11.40 | $0.228 | |
| $3.00 | $15.00 | $0.3 | N/A | N/A | — | — | — | — | — | — | — | — | — | — | — | |
| $2.50 | $12.50 | $0.25 | N/A | N/A | $1.25 | $6.25 | $0.15 | N/A | — | — | — | — | — | — | — | |
| $2.83 | $14.13 | $0.28 | N/A | N/A | — | — | — | — | — | — | — | — | — | — | — | |
| $3.60 | $18.00 | N/A | N/A | N/A | — | — | — | — | — | — | — | — | $1.80 | $9.00 | N/A | |
| $3.00 | $15.00 | $0.3 | N/A | N/A | — | — | — | — | — | — | — | — | — | — | — | |
| $3.00 | $15.00 | $0.3 | N/A | N/A | — | — | — | — | — | — | — | — | — | — | — | |
| $3.00 | $15.00 | $0.3 | N/A | N/A | — | — | — | — | N/A | N/A | N/A | N/A | — | — | — | |
| — | — | — | — | — | — | — | — | — | $4.50 | $22.50 | $0.45 | N/A | — | — | — | |
Cost Calculator
US Dollar ($)
Preset:
Cheapest Instances to Run It
Cloud GPU instances that can host Kimi K3, ranked by cheapest on-demand price. The model needs about 6672 GB of GPU memory at FP16 precision (estimated from its parameter count), so treat the fit as guidance rather than a guarantee.
All clouds
FP16 (full precision)
US Dollar ($)
No GPU instances fit Kimi K3 at FP16 precision — try a lower precision such as INT4.
Versions
| Version | Released | Context | Input / 1M | Output / 1M | Status |
|---|---|---|---|---|---|
| Kimi K3.1 Pro | 1.0M | $2.00 | $10.00 | Available | |
| Kimi K3.1 Pro Highspeed | 1.0M | $4.00 | $20.00 | Available | |
| Kimi K3 | 1.0M | $0.192 | $12.50 | Current | |
| Kimi K3 Batch | — | 1.0M | — | — | Available |
| Kimi K3 Coding | — | 1.0M | $0.440 | $1.61 | Available |
| Kimi K3 Fast | — | 1.0M | $4.50 | $22.50 | Available |
| Kimi K3 P | — | — | — | — | Available |
| Kimi K3 Spot | — | 1.0M | $2.10 | $10.50 | Available |
| Kimi K3 US | — | 1.0M | $4.50 | $22.50 | Available |
| Kimi K2.7 Code HighSpeed | 262K | $1.90 | $8.00 | Available | |
| Kimi K2.7 Code | 262K | $0.380 | $1.60 | Deprecated |