Loading...
Loading...
Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It is optimized for agentic capabilities, including advanced tool use, reasoning, and code synthesis. Kimi K2 excels across a broad range of benchmarks, particularly in coding (LiveCodeBench, SWE-bench), reasoning (ZebraLogic, GPQA), and tool-use (Tau2, AceBench) tasks. It supports long-context inference up to 128K tokens and is designed with a novel training stack that includes the MuonClip optimizer for stable large-scale MoE training.
Quality Index
26.3
130th of 444
Top 30%
Coding Index
22.1
131st of 354
Top 38%
Math Index
57.0
123rd of 268
Top 46%
Price/1M
$1.00
493rd cheapest
234% above median
Top 72%
Speed
37 tok/s
Top 55%
TTFT
0.93s
Context Window
131K
145th largest
Top 63%
Input
$0.57
per 1M tokens
Output
$2.40
per 1M tokens
Blended
$1.00
per 1M tokens
Cheaper than 28% of models. Median price is $0.30/1M tokens.
Daily
$1.00
Monthly
$30.06
37
tokens/sec
Faster than 45% of models
0.93
seconds
Faster than 33% of models
0.93
seconds
Faster than 40% of models
Market Median
45 tok/s
19% slower
Median TTFT
0.42s
123% slower
Throughput/Dollar
37
tok/s per $/1M
Speed Comparison
Context Window
131K
tokens
Larger than 37% of models
Max Output
131K
tokens
100% of context