Kimi K3.1 Pro: Intelligence, Performance & Price Analysis
Artificial Analysis's model page for Kimi K3.1 Pro, covering its Intelligence Index score, output speed, latency, pricing and context window as measured in October 2026.
Summary of the Artificial Analysis model page, as read on October 23, 2026. Original: artificialanalysis.ai/models/kimi-k3-1-pro
Intelligence
Artificial Analysis Intelligence Index (v4.3.2): 55, ranked #4 of 215 models in its comparison class, against a median of 27
Reasoning model; text and image input, text output
The index run used about 150M output tokens (median 87M) and cost about $2.11 per index task
Speed and latency (Kimi's API)
Output speed: 73.5 tokens per second
Time to first token: 3.20 seconds
Price and context
$2.00 per 1M input tokens and $10.00 per 1M output tokens, with a 90% cache discount
1-million-token context window
Notes
At the time of reading, the page listed the model as proprietary with weights not yet public (Moonshot has promised weights by November 4, 2026), and its FAQ gave a size of 2.8 trillion parameters, where Moonshot itself says only "3T-class".