Kimi K2 0711
by Moonshot AIRank #224Score 51.4
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...
Performance Overview
Score
51.4
Rank
#224
24h Change
0
7d Change
-63
State
stableConfidence
highSignal Scores
| Signal | Normalized | Weight | Contribution | Freshness |
|---|---|---|---|---|
Capabilities capability | 33.3 | 20% | 6.7 | 2026-08-08T12:28:34.420Z |
Benchmarks benchmark | 50.7 | 30% | 15.2 | 2026-08-08T12:28:34.420Z |
Pricing pricing_tier | 97.7 | 15% | 14.7 | 2026-08-08T12:28:34.420Z |
Context Window context_window | 81.2 | 10% | 8.1 | 2026-08-08T12:28:34.420Z |
Recency recency | 61.3 | 15% | 9.2 | 2026-08-08T12:28:34.420Z |
Output Capacity output_capacity | 83.1 | 10% | 8.3 | 2026-08-08T12:28:34.420Z |
Top Drivers
neutral
Benchmarks
positive
Pricing
$2.30/M output tokens
$2.30
positive
Recency
Released over 1 year ago
1.1y ago
positive
Output Capacity
Up to 100K output tokens per request
100K
Capabilities
| Capability | Supported |
|---|---|
| Vision | No |
| Reasoning | No |
| JSON Mode | No |
| Streaming | Yes |
| Function Calling | Yes |
| Web Search | No |
Pricing
Input / 1M tokens
$0.57
Output / 1M tokens
$2.30
Context Window
131K
Max Output
100K