kimi-k2.6 vs kimi-k3
Side-by-side comparison of kimi-k2.6 and kimi-k3: benchmarks, pricing, context window and capabilities. Both are accessible through Requesty's unified API. kimi-k3 outperforms kimi-k2.6 on 4 of 4 shared benchmarks.
kimi-k2.6
Input / 1M
$0.95
Output / 1M
$4.00
Context
262K
Model ID
moonshot/kimi-k2.6
kimi-k3
Input / 1M
$3.00
Output / 1M
$15.00
Context
1.0M
Model ID
moonshot/kimi-k3
Benchmark comparison
Intelligence Indexreasoning
kimi-k2.635.8%
kimi-k350.2%
Coding Indexcoding
kimi-k2.661.8%
kimi-k376.2%
GPQA Diamondreasoning
kimi-k2.691.1%
kimi-k393.5%
Terminal-Bench Hardagentic
kimi-k2.643.9%
kimi-k3N/A
τ²-Benchagentic
kimi-k2.695.9%
kimi-k3N/A
SciCodecoding
kimi-k2.6N/A
kimi-k359.5%
Humanity's Last Examreasoning
kimi-k2.637.5%
kimi-k346.9%
Scores sourced from official model cards, Artificial Analysis, and public leaderboards. Benchmarks measure specific skills and don't capture every aspect of model quality.
Pricing & specifications
| kimi-k2.6 | kimi-k3 | |
|---|---|---|
| Input price / 1M | $0.95 | $3.00 |
| Output price / 1M | $4.00 | $15.00 |
| Context window | 262K tokens | 1.0M tokens |
| Max output | 262K tokens | 1.0M tokens |
| Vision input | Yes | Yes |
| Tool calling | Yes | Yes |
| Reasoning | Yes | Yes |
| Prompt caching | Yes | Yes |
| Computer use | N/A | N/A |
| Provider | Moonshot AI | Moonshot AI |
Questions people ask
Is kimi-k2.6 better than kimi-k3?
kimi-k3 outperforms kimi-k2.6 on 4 of 4 shared benchmarks. See the benchmark comparison above for specifics: kimi-k2.6 and kimi-k3 have different strengths across reasoning, coding, math and multimodal tasks.
Which is cheaper, kimi-k2.6 or kimi-k3?
kimi-k2.6 is cheaper. kimi-k2.6 costs $0.95/$4.00 per 1M input/output tokens, while kimi-k3 costs $3.00/$15.00.
Can I use kimi-k2.6 and kimi-k3 through the same API?
Yes. Requesty provides a single OpenAI-compatible API that routes to both. Change just the "model" parameter to switch between "moonshot/kimi-k2.6" and "moonshot/kimi-k3", no other code changes needed.
What are the context windows?
kimi-k2.6 supports up to 262K tokens of context. kimi-k3 supports up to 1.0M tokens. Longer context means you can feed larger documents or codebases in a single prompt, though quality often degrades past 128K for most models.
Switch between kimi-k2.6 and kimi-k3 with one line of code
Requesty provides a single OpenAI-compatible API for 600+ models. Change the model parameter, not your code.