Umans Kimi K3.1 Pro
64.6tok/s
throughput · p50 · last 5 min
1.21s
TTFT · p50 · last 5 min
100.00%
uptime · 24h
Kimi K3.1 Pro: Moonshot's newest open model and the first of the K3.1 series, post-trained from Kimi K3 and served from the official Kimi-K3.1-Pro open weights. A 2.8T mixture-of-experts (104B active per token) with native image understanding and a 1M-token context window, which Moonshot reports as substantially stronger than K3 on coding, factual accuracy and agentic work. It thinks by default at maximum reasoning effort; select none, low, high or max to trade depth for speed.
Context
1049K
Max output
131K
Recommended
131K
Vision
Yes
Tools
Yes
Reasoning
Toggle · none/low/high/max
Weights
Trends
Speed over the last 90 days
90 days agopre-release before Nov 6, 2026today
90 days agopre-release before Nov 6, 2026today
Changelog
Events for Umans Kimi K3.1 Pro
Nov 62026
Released pay-per-token: Umans Kimi K3.1 Pro Released
umans-kimi-k3.1-pro joins the lineup: Kimi-K3.1-Pro, the checkpoint that has served here as a seat-gated Labs experiment since November 4. Moonshot's newest open model, post-trained from Kimi K3: a 2.8T mixture-of-experts (104B active per token) with native image understanding and a 1M-token context window. It thinks by default at maximum reasoning effort; none, low, high and max trade depth for speed. Billed per token: $2.00 / $10.00 / $0.20 per 1M (input / output / cache read). The experiment's metrics stay on the model's status page as its pre-release period. The Labs experiment on umans-kimi-k3.1-pro-lab is unaffected.